Skip to main content

Full text of "Outlines of industrial chemistry; a text-book for students"

See other formats




















All rights reserved 

COPYRIGHT, 1898, 1905, AND 1916, 

Set up and electrotyped. Published November, 1898. Reprinted, 
with corrections, November, 1899. 

New edition, corrected and enlarged, May, 1905; April, 1907; 
January, October, 1908; August, 1909; April, 1911; September, 
1912; August, 1913; September, 1914. 

Third edition, revised, May, 1916. Reprinted November, 1916. 


J. 8. Gushing Co. Berwick & Smith Co. 
Norwood, Mass., U.S.A. 

3E0 tfje fHemxjrjj ot 


<Tf)ts Book is DrttratetJ 






THE object of this book is to furnish, an elementary course in 
Industrial Chemistry, which may serve as the ground work for 
a more extended course of lectures, if desired. The writer has 
endeavored to describe briefly, within the limits of one moderate- 
sized volume, the more important industrial chemical processes, but 
omitting matters of detail which properly belong in the larger hand- 
books. Numerous references are made in the text to periodicals 
and journals, and to the standard handbooks and encyclopaedias 
and many special works, where details, lacking in this book, may 
be found. The bibliographical lists following each section are not 
complete, but only include those works which will usually be found 
in most chemical libraries ; the references to the journal literature 
are merely those articles to which the author's attention has been 
drawn in the preparation of class-room exercises. The diagrams 
illustrating the text have, in most cases, been drawn as simply as 
possible, purposely showing only the essential features. 

In the selection and order of arrangement of the several sub- 
jects, the author has necessarily been influenced by his work in this 
Institute and the requirements of his own class, but it is believed 
that the book as a whole will be found applicable to the work in 
most institutions of learning where industrial chemistry is taught. 
The subject of Metallurgy has been entirely omitted, since there are 
already several excellent brief text-books dealing with it alone, and 
instruction in it is generally given independently of that relating to 
technical chemistry. Likewise the important subject of the coal-tar 
colors has been condensed into the briefest possible outline, because 
this is nearly always included in courses in organic chemistry, and 
there are several small manuals treating of it. Analytical processes 
have also been omitted as foreign to the intended scope and purpose 
of the book. 

It is assumed that students taking this course are familiar with 
the elements of general chemistry, both inorganic and organic, and 
with the elements of physics. 

In the compilation of this work, free use has been made of many 



of the standard English, German, and French hand-books and ency- 
clopaedias, particularly of Professor T. E. Thorpe's Dictionary of 
Applied Chemistry, the works of Professor Lunge on Sulphuric 
Acid and Alkali, and Coal-tar and Ammonia, Ost's Technischen 
Chemie, and Dainmer's Handbuch der chemischen Technologic. 

The following business firms have courteously loaned cuts and 
drawings for the illustrations : Curtis Davis & Co., Cambridgeport, 
Mass., slabbing machine and soap kettle ; William Campbell & Sons, 
Cambridgeport, Mass., rendering tank ; The De La Vergne Refriger- 
ating Machine Co., New York, refrigerating machine ; H. L. Dixon, 
Pittsburg, Pa., tank furnace for glass; United Gas Improvement 
Co., Philadelphia, water gas plant ; John Johnson & Co., New York, 
filter press ; Sernet-Solvay Company, Syracuse, N.Y., coke oven ; 
R. D. Wood & Co., Philadelphia, Pa., Taylor gas producer. 

The writer wishes to acknowledge his indebtedness to the fol- 
lowing friends for assistance and advice in the revision of those 
portions of the work treating of their specialties : C. D. Jenkins, 
State Gas Inspector of Massachusetts, Illuminating Gas ; A. D. 
Little, Consulting Chemist, Wood Pulp and Paper ; J. W. Loveland, 
Superintendent Curtis Davis & Co., Soap Manufacturers, Soap, 
Candles, and Glycerine; F. G. Stantial, Superintendent Cochrane 
Chemical Co., Sulphuric, Hydrochloric, and Nitric Acids ; the fol- 
lowing members of the instructing staff of the Massachusetts Insti- 
tute of Technology : Professor A. H. Gill, Fuels and Oils ; G. W. 
Rolfe, Starch, Glucose, and Sugar; S. C. Prescott, Fermentation 
Industries ; J. W. Smith, Textile Industries. 

Special thanks are due to Dr. W. R. Whitney, of the Institute of 
Technology, for much assistance in the proof-reading, and also to 
Dr. B. L. Robinson, Curator of the Gray Herbarium, Harvard Uni- 
versity, for his painstaking revision of the botanical nomenclature. 

In the labor of preparation of the book, the author has also had 
much help from his wife, who copied the entire manuscript and has 
assisted in the reading of all of the proof. 

BOSTON, MASS., October, 1898. 

In this new edition, such errors as have been brought to the 
writer's notice have been corrected, but material changes have not 
been generally attempted, owing to press of other work. The author 
wishes to express his indebtedness to those who have called attention 
to weak parts in the book, and heartily invites further criticism. 

F. H. T. 

BOSTON, MASS., October, 1899. 


THE important advances made in the chemical industries since 
the appearance of the first edition, and the establishment of several 
successful manufactures which at that time were not beyond the 
experimental stage, have necessitated giving the text considerable 
revision, with material additions and corrections. Such errors of 
statement or of proof-reading as have been noted have been cor- 
rected, and further criticism by instructors and others interested 
in the subject is invited. 

It appeared desirable that a short outline- of elementary metal- 
lurgy should be included, in order that the book might better meet 
the requirements of the courses of study of certain colleges and 
technical schools. With the exception of the paragraphs upon 
bismuth, cadmium, and magnesium, this has been prepared by Mr. 
Chas. D. Demond, and is now included as Part III of the new 

In connection with the material introduced, the writer wishes 
to express here his obligations to the following firms for permission 
to make abstracts and copy illustrations from their publications 
and catalogues : The Allis-Chalmers Co., Chicago ; The Engineering 
and Mining Journal, New York ; The Sugar Apparatus Manufactur- 
ing Co., Philadelphia. 

F. H. T. 

BOSTON, April, 1905. 



THE great progress which has been made in Chemical Industry 
since the publication of the second edition of this work in 1905 has 
necessitated entire rewriting of many sections of the book, with 
elimination of much obsolete matter and the introduction of much 
new material. While the general plan of the former editions has 
been retained, in treating the various subjects use has been made of 
the modern concepts and theories of chemistry wherever these 
promised to make clearer the phenomena involved. Extended math- 
ematical or theoretical discussion of the processes has been avoided 
as beyond the scope of the book. It is not supposed that the expert 
will find this book a guide in his own particular field, for its purpose 
is to impart to students and others not already familiar with the 
processes of chemical industry, some knowledge of the plant and 
methods employed in the more important manufacturing operations 
based upon chemical changes. 

F. H. T. 

W. K. L. 
BOSTON, March, 1916. 




Objects of Industrial Chem- 




Spontaneous .... 
By direct heat . . . , 
By steam heat .... 

In vacuum 

Vacuum pans . . . 
Multiple effect systems 
Yaryan evaporator . 
Kestner evaporator 
Lillie evaporator 


Fractional condensation 
Dephlegmation .... 
Coupler's still .... 
French column apparatus 

Coffey still 



Bag niters 

Suction filtration .... 
Pressure filtration, with the 
filter press . . . ':-._ 
By use of leaf filter ... 
Centrifugal filtration . . 
Sand filters . ... , 



Reverberatory furnace . . 
Revolving furnace . . . 
Muffle furnace .... 
Shaft furnace or kiln . 


























Refrigeration 23 

Compression machines . .23 

Absorption machines . . 23 

Chilling by compressed air 25 

Specific Gravity ..... 25 

Hydrometers 25 

Pyknometer 27 

Westphal's balance . . . 28 

Surface Phenomena and 

Colloids 28 

Disperse systems .... 28 

Colloids 30 


Solid fuels 32 

Wood, peat, lignite, brown 
coal, bituminous coal, 
anthracite, charcoal, 

coke 32-37 

Beehive coke oven . . 35 

By-product coke ovens . 36 

Liquid fuels 38 

Crude petroleum and oil 

residues 38 

Gaseous fuels 38 

Natural gas ....'. 38 

Coal gas 39 

Water gas 39 

Producer gas 41, 

Siemens' gas producer . 42 

Taylor's gas producer . 42 

Mond's gas process . . 43 

Blast furnace gas ... 43 


Sources of natural waters . 46 

Impurities of natural waters . 46 





Hard, soft, saline, alkaline . 47 
Purification by chemical pre- 
cipitation 48 

Boiler scale 50 

Classification of boiler 

waters for locomotives . 52 
Water for various special in- 
dustries 52 


Extraction from its ores . . 55 

Recovered sulphur .... 57 

Sulphur derivatives . . 58-61 

Sulphur dioxide .... 58 

Sodium bisulphite ... 59 

Calcium bisulphite ... 60 

Hydrosulphurous acid ... 60 

Sodium hydrosulphite . . 60 

Sodium thiosulphate . . 61 


Commercial grades of acid . 62 
Theories of the formation of 

the acid in chamber process 63 
Materials for manufacture of 

the acid 66 

Brimstone 66 

Pyrites . 66 

Pyrites burners for lump 

ore 67 

Pyrites burners for 

''fines" 68 

Glover tower 70 

Lead chambers .... 70 
Gay-Lussac tower . . .71 

Acid egg . 73 

Kestner's elevator . . . 73 

Air-lift elevator for acid . 74 

Concentration of acid . . 74 
Porcelain and fused silica 

vessels 74 

Glass and platinum stills 75 

Cast-iron stills .... 76 

Kessler's apparatus . . 76 

Lunge's plate tower ... 77 

Barbier's tower system . . 78 

Catalytic processes for acid 

making 79 

Purification of gases for 

contact process ... 81 
Fuming sulphuric acid . . 82 


Sources of salt 83 

Preparation of salt by various 
processes 84-87 


Salt-cake furnaces .... 88 

Open roaster 88 

Muffle roaster 89 

Coke tower for acid absorp- 
tion 90 

Bombonnes or tourills for acid 

condensation ..... 90 

Hargreaves-Robinson process 

for sodium sulphate ... 92 

Sodium sulphate or salt-cake 92 


Leblanc process 94 

Black-ash or balling fur- 
nace 95 

Lixiviation of black-ash . 97 
Carbonation, purification, 
and evaporation of tank 

liquor 98 

Thelan's pan 100 

Soda crystals or sal soda . 100 

Caustic soda 101 

Loewig's process . . . 102 

Tank waste 103 

Methods of treating tank 

waste .... 104-107 
Ammonia soda process . . 107 
Carbonating tower . . . 108 
Parnell-Simpson modifica- 
tion of this process . .111 
Frasch process for caustic 

soda HI 

Cryolite soda process . . .113 


Processes using manganese 

oxides 115-118 

Dunlop's method . . .117 
Weldon's process . . . .117 



Deacon's process with copper 
salts 118 

Processes employing nitric 
acid 121 

Magnesia processes for chlo- 
rine 122 

Weldon-Pechiney process . 123 

Processes for recovering 
chlorine from ammonia- 
soda waste liquor . . . 123 

Electrolytic Processes for 
chlorine and caustic 

soda 124 

Le Sueur's process . . .126 
Carmichael's apparatus . 126 
Hargreaves-Bird apparatus 126 

Townsend cell 127 

Griesheim-Elektron process 127 
Castner's process . . .128 

Whiting's cell 128 

Bell's apparatus .... 129 
Rhodin's apparatus . . . 129 
Gravity cell . . . .; . 129 
Acker process 130 

Hypochlorites T . . . . 131 
Bleaching powder and 
bleach liquors . . 131-132 

Chlorates . . . . . ." .'135 

Perchlorates . . .,. . . 135 

Persulphates . . . . . .136 


Methods of manufacture . . 137 
Cylindrical iron retorts for 

nitric acid 138 

Guttmann's apparatus . . 138 

Hart's apparatus .... 139 

Valentiner's process . . .140 

Rhenania process . . .140 

Fuming nitric acid .... 141 

Nitric acid from the nitrogen 

of the air 142 

Bradley and Lovejoy's 

process 142 

Birkeland and Eyde 

process 142 

Schoenherr process . . . 143 

Pauling process .... 144 

Commercial nitrates 145-149 



Sources of ammonia . . . 150 

Synthetic preparation of am- 
monia 150 

Frank and Caro process . 150 
Haber's process .... 150 
Serpek's process .... 151 

Ammonia from gas liquor . . 151 
Feldmann's apparatus . .152 
Griineberg-Blum apparatus 152 

Ammonia from distillation of 
waste animal matter . .153 

Distillation of peat as source 
of ammonia . . . . . 153 

Ammonium salts of com- 
merce 154-155 


Sources of potassium salts . 156 
Potash from wood ashes . 156 
Potash from beet sugar 

molasses 157 

Potash from wool scourings 157 
Potash from sea- weeds . .158 
Stassfurt deposit of potas- 
sium salts 158 

Potassium salts of commerce 162 


Requisites of a fertilizer . . 164 

Waste materials as sources of 

fertilizer products . . . 165 
Blood, bones, garbage, 
"tankage," etc. . 165-166 

Peruvian and fossil gua- 
nos 166-167 

Phosphate rocks 167 

Apatite 167 

Phosphorites .' * '*. . . 168 

Superphosphates ., ". . 169 
Reverted phosphate . . .170 

Phosphatic slag . . . , .171 

Gypsum or "plaster" as 
fertilizer 173 

Sewage as fertilizer . . .173 


Lime 175 

Properties of Lime . . .175 




Lime burning 175 

Limekilns . . . 175-177 
Hydraulic lime . . . .178 

Mortar 179 

Sand-lime bricks .... 180 

Cements 181 

Manufacture of cement 182-189 
Kilns for burning ce- 
ment .... 185-187 
Mills for cement grind- 
ing 187-189 

Constitution of Portland 

cement 190 

Hardening of cement . .190 

Testing of cement . . .191 

Plaster of Paris 193 


Properties and composition 

of glass 196 

Lime and lead glass . . .197 
Materials for glass making 197 
Glass furnaces . . 199-201 
Glass pots, open and closed 201 
General process of glass 
making .... 202-203 

Plate glass 204 

Window glass and glass blow- 
ing 205 

Crown glass 206 

Cut glass and pressed ware . 206 

Tempered glass 207 

Compound glass ..... 207 
Colored glass .... 207-209 

Enamel . 209 

Iridescent glass 209 

Mirrors 209 


Kaolin or china clay . . .212 
Fire-clay, pipe- or ball-clays 213 
Empirical and rational an- 
alyses of clays . . . .214 

Ceramics 215 

Non-porous ware . . .215 

Porcelains 215 

Stoneware 216 

Kilns . . . 217 


Porous ware 217 

Faience and common 

pottery 217 

Tiles . 218 

Vitrified, encaustic, 

and glazed tiles . . 218 
Glazes 219 

Crazing of glaze . .219 

Terra cotta 220 

Bricks 220 

Fire-brick .... 221 


White pigments . . . 223-231 

White lead 223 

Dutch process .... 223 

Chamber process . . . 225 

Carter's process . . . 226 

Thenard's process . . 226 

Electrolytic processes . 227 

Liickow process . . 228 

Substitutes for white lead 229 

Sublimed white lead . 229 

Lead sulphite . . . 229 

Pattinson's white lead 229 

White zinc or Chinese 

white . . ' 229 

.Lithopone 230 

'Barytes 230 

Gypsum, terra alba . . .231 
Whiting or Paris white . .231 

China clay 231 

Blue pigments . . . 231-235 

Ultramarine 231 

Prussian or Berlin blue . 233 

Smalt 234 

Cobalt blue 235 

Copper blues 235 

Indigo 235 

Green pigments . . . 236-238 
Ultramarine green . . . 236 
Brunswick green .... 236 
Chrome greens .... 236 
Guignet's green . . . .237 
Copper greens, malachite 

and verdigris .... 237 
Copper-arsenic greens . . 238 
Scheele's and Paris 
greens . . . . . . 238 




Terra verde 238 

Yellow pigments . . 238-241 
Chrome yellows .... 238 
Yellow ochre and siennas . 240 
Cadmium yellow .... 240 
Orpiment or royal yellow . 240 

Litharge 241 

Gamboge 241 

Indian yellow or purree . .241 

Orange pigments .... 241 
Orange mineral . . . .241 
Antimony orange . . . 242 

Red pigments . . . 242-246 

Red lead 242 

Chrome red or American 

vermilion 243 

Red ochre, Indian red, light 

red 243 

Iron reds, Venetian red,, 
rouge, colcothar . . . 243 

Vermilion 244 

Realgar and antimony reds 245 
Carmine and " lakes " . . 245 

Brown pigments 246 

Umbers, Vandyke brown . 246 
Sepia 247 

Black pigments . . . 247-248 


Sources, and methods of ex- 
traction from them . . . 249 
Bromides 251 


Extraction from kelp and 
varec 252 

Recovery from mother- 
liquors of sodium nitrate 
industry . 253 

Iodides . . . . ' . . . .254 


Preparation from bone-ash . 256 
Preparation from mineral 

phosphates 256 

Readman's electric furnace 

process for reduction . . 257 
Matches . 258 


Sources and preparation . . 260 

Borax 261 

Perborates 263 


Carborundum 264 

Artificial graphite . . . .265 

Calcium carbide 266 

Calcium cyanamide . . . 267 

Alundum . 267 

Barium hydroxide .... 268 
Cyanides 268 


Arsenious acid, white arsenic 269 

Arsenic acid 269 

Arsenates, sodium and lead 270 
Arsenites, sodium . . . 270 



Barium peroxide 
Hydrogen peroxide 
Sodium peroxide 



Preparation from potassium 

chlorate 275 

Boussingault-Brin process of 

preparation 275 

Deville's process . . . . . 276 
Tessie du Motay process . . 276 
Linde refrigeration process . 277 
By electrolysis of water . . 277 


Ferrous sulphate, green vit- 
riol, copperas 279 

Copper sulphate, blue vitriol, 

bluestone 280 

Zinc sulphate, white vitriol . 281 
Aluminum sulphate, from 

clay and from bauxite . 282 
Bayer's process for pure 

alumina 283 

Aluminum sulphate from 

cryolite 284 

Alum 285 

Preparation fromalunite . 286 



Preparation from alum 

shales or slate .... 286 
"Neutral alum" .... 287 

Sodium alum 287 

Iron alums and chrome 
alum 288 


Preparation from ammonia 
and carbon at high tem- 
peratures 289 

Recovery of cyanide from 

coal gas by Bueb's process 289 
Bunsen-Playfair process of 

preparation 290 

Raschen's process .... 290 
Ammonium sulphocyanide 
from carbon disulphide 
and ammonia by Gelis 
process 290 


Recovery from spent iron 

oxide of gas purifiers . .291 
Potassium ferrocyanide, re- 
covery from spent iron 

oxide 291 

Preparation from waste 

nitrogenous matter . . 292 
Potassium ferricyanide, red 

prussiate of potash . . . 293 

Potassium cyanide .... 294 

Beilby's process .... 295 

Castner's process . . . 295 



. 298 

NATES ... . 299 





Pyroligneous acid . . . .301 
Kilns and retorts for wood 

distillation 302 

Methyl alcohol or wood spirit 305 

Acetone 305 

Acetic acid 306 

Acetates 308 

Wood-tar 309 

Creosote oil 310 

Stockholm tar .... 310 


Bone oil 311 

Bone-black 311 


Carburetted water-gas . . 312 

Coal-gas 314 

Plant for distilling coal for 
gas 315 

Purification of gas . . . 320 
Feld process of purifica- 
tion 321 

Recovery of cyanide 
from coal-gas . . . 322 

Oil gas 323 

Blau gas 324 

Acetylene 324 

Air gas 325 


Properties of tar 327 

Distillation of tar . . . . 327 

First runnings 330 

Light oil 330 

Naphtha 330 

Carbolic oil 331 

Creosote oil ..... 332 

Naphthalene 332 

Anthracene oil .... 332 

Pitch 333 

Yields of crude and pure 

products from tar . . . 333 




Petroleum industry . . . 334 
Distribution and origin of 

petroleum 334 

Oil-well drilling . . . .336 

Crude petroleum .... 338 

Refining of petroleum . . 339 

' ' Cracking ' ' of heavy oils . 340 

Purification of distillates . 341 

Burning oils .... 342 

Paraffine oils .... 342 

"Neutral oils" ... 342 

Spindle oils, machinery 

oils, cylinder oils . . 343 
Reduced oils . . . .343 

Vaseline 343 

Russian petroleums . . . 343 
Oil testing . . . . . .344 

Shale oil industry .... 345 

Ozokerite 346 

Asphalt 347 



Properties of the fatty oils . 349 
Hydrolysis of fats . . . .351 
Occurrence and extraction of 

the vegetable oils . . . 352 
Occurrence and extraction of 

the animal oils . ,. . . 354 
Testing of fatty oils . . .355 
Classification of oils . . . 356 
Vegetable drying oils . . 357 
Vegetable semi-drying oils 359 
Vegetable non-drying oils . 362 
Marine animal oils . . . 363 
Terrestrial animal oils . . 365 
Solid vegetable fats . . . 366 
Solid animal fats .... 367 

Waxes 368 

Liquid waxes 368 

Solid animal waxes . . . 369 
Solid vegetable wax . . . 370 


Saponification 372 

Soap kettles 374 

Cold process soap .... 374 
Boiled soaps 375 

Yellow (rosin) soaps 
"Boiled down soaps" 
Toilet soaps . . . 
Milled soaps . . 
Remelted soaps 
Transparent soaps 
Scouring soaps . . 
Soap powder . 


. 375 
. 377 

. 378 
. 378 
. 378 
. 378 
. 379 
. 379 


Dipped, poured, and moulded 

candles 380 

Saponification of fats for 
candle stock 381 


Van Ruymbeke process for 
recovery of glycerine from 
spent soap lyes .... 384 

Glycerine from candle stock . 385 

Glatz process for glycerine 
from soap lyes .... 385 

Properties and uses of glyc- 
erine 386 


Properties, and methods of 

extraction, 387 

Characteristics of the in- 
dividual essential oils 388-392 


Resins 393-396 

Varnishes, spirit, turpen- 
tine, and linseed oil var- 
nish 397 

Oleo-resins 398 

Balsams 398 

Gum-resins, properties of the 
individual gum-resins 398-399 

Gums, properties of the in- 
dividual gums . . . 399-400 

Occurrence and properties of 

starch. . . , v . . . 401 

Corn starch 402 

Wheat starch 407 

Potato starch 408 

Rice starch . . . . .409 
. 410 




Arrowroot 410 

Cassava 411 

Dextrin 412 

Manufacture of dextrin and 
British gum .... 412 

Glucose 412 

Dextrose, levulose, and 

commercial glucose . 413 

1 Conversion 414 

Neutralization .... 415 
Bone-char nitration . . 416 


Occurrence and properties of 

cane sugar 420 

Manufacture of raw sugar 

from sugar cane . . .421 
Manufacture of raw sugar 

from sugar beets . . . 425 
Sugar refining 428 


Fermentation 435 

Organized ferments, mould 
growths, bacteria, yeast 435 

Wine 440 

Composition of grape juice 440 
Extraction of the must . . 440 
Fermentation of the must . 441 
Clarification and preserva- 
tion of the wine . . . 442 
"Improving" of the new 
wine . .... . . 443 

Champagne ..... 443 

Other wines 444 

Brewing 444 

Malting 445 

Steeping, couching, and 

flooring 446 

Pneumatic malting . . 447 

Mashing 448 

Infusion method . . . 449 

Decoction method . . 450 

Boiling of the wort . . . 451 

Hops 451 

Cooling of the hot wort . 452 
Fermenting of the wort . . 452 
' ' Vacuum process ' ' of 
fermentation . . 454 


Extract in beer .... 454 

Bottling or barrelling . . 454 

Brewed liquors .... 455 

Distilled liquors 456 

Manufacture of alcohol . 456 
Distillation of the fer- 
mented mash . . . 458 
Purification and recti- 
fication of raw spirit . 459 
Revenue restrictions 

upon the industry . . 459 
Denatured alcohol, 

" methylated spirit " . 460 

Fusel oil 461 

Whiskey 461 

Gin 462 

Brandy 462 

Rum . 463 

Liqueurs, cordials, arrack, 

absinthe 463 

Vinegar 463 

Orleans process of manu- 
facture 464 

" Quick " vinegar proces^ . 465 
Cider, wine, malt, and 

spirit vinegars .... 466 
Lactic fermentation and 

lactic acid 467 


Characteristic properties of 

explosives 470 

Gunpowder 471 

Pebble and prismatic 

powders 474 

Brown or cocoa powder . 475 

Mining powders . . . 475 

Nitrocellulose or guncotton 475 

Pyroxyline ..... 479 

Nitroglycerine .... 479 

Dynamite . . .. . .481 

Explosives with an active 

"dope" 482 

Forcite 483 

Blasting gelatine and 

gelatine dynamite . . 483 

Smokeless powders . . 483 

Picrates and picric acid . . 484 



Fulminates and fulminic 
acid 484 

Azides of heavy metals as 
detonators 484 

Sprengel explosives . . . 485 

Military explosives . . . 485 
Melinite, lyddite, shi- 
mose, trinitrotoluol . 485 


Fibres .487 

Vegetable fibres . . . .487 

Cotton fibre 487 

Mercerized cotton . . 489 
Alkali cellulose, "vis- 
cose" 490 

Linen 490 

Hemp 491 

Jute 491 

China grass (ramie] . . 492 

Esparto 492 

Manila, sisal, and sunn 

hemp 492 

Cocoanut fibre .... 492 

Animal fibres 492 

Silk 492 

Artificial silk . . .496 

Wool 497 

Wool scouring and re- 
covery of wool 

grease 499 

Carbonizing of vege- 
table fibre in wool 501 

Bleaching 501 

Cotton bleaching . . .501 
Madder bleach for calico 

print cloth .... 503 
Turkey-red bleach for 
cotton to be dyed with 

alizarins 506 

Market bleach for com- 
mercial white goods . 506 
Mather-Thompson pro- 
cess for bleaching . . 507 
Hermite bleaching pro- 
cess 508 

Hydrogen peroxide and 
permanganates as cot- 
ton bleaches . . 508 

Linen bleaching .... 508 
Irish process .... 508 
Jute bleaching .... 509 
Hemp bleaching .... 509 
Wool bleaching . . . .510 
Stretching of yarn be- 
fore scouring and 

bleaching 510 

"Crabbing" of union 

goods 510 

Stoving 511 

Hydrogen peroxide 

bleach 512 

Silk bleaching 512 

Mordants 512 

Metallic mordants . . . 513 
Organic mordants, . . . 518 

Tannins 518 

Coloring matters . . . .521 
Natural dyestuffs . . .521 
Artificial dyestuffs . . . 526 
Relation of color to con- 
stitution 527 

Dyeing .528 

Theories of the dyeing 

process 529 

Methods of dyeing textiles 530 
Grouping of commercial 
dyes according to 
method of application 
to the fibre . . . .531 

Direct dyes 532 

Basic dyes 533 

Acid dyes 535 

Mordant dyes .... 536 
Acid-mordant dyes . . 539 
Sulphide dyes .... 540 

Vat dyes 541 

Ingrain colors .... 543 

Textile printing 546 

Block printing .... 546 
Machine printing . . . 547 
Color mixing . . . . . 548 

Styles 549 

Pigment style .... 549 

Steam style 549 

Madder style .... 550 
Oxidation style . . . 550 




Discharge style . . . 550 

Resist style 551 

Wool printing 551 

Silk printing 551 


Materials for paper . . . 554 

Wood pulp . . . . . . 554 

Mechanical pulp . . . 554 

Chemical pulp . . . 555 

Soda process . . . 555 

Sulphite process . . 555 

Sulphate process . . 558 

Rags 561 

Esparto 561 

Jute 562 

Bleaching of paper pulp . . 562 
Paper making process . . . 562 

Furnishing 563 

Sizing 563 

Hand-made paper . . . 564 

Cylinder machine . . . 564 

Fourdrinier machine . . 564 

Printing paper . . . 565 

Wrapping paper . . . 565 

Writing paper .... 565 

Blotting and tissue 

papers 565 

Parchment paper . . . 565 
Willesden paper . . . 566 
Vulcanized fibre . . . 566 


Colloidal characteristics of 

glue 568 

Sources of glue 568 

Preparation of glue . . . 568 

Hide glue 568 

Bone glue 570 

Fish glue 570 

Liquid glue 570 

Gelatine 570 

Isinglass 571 

Vegetable gelatine (agar agar) 571 


Structure of the skin . . . 572 
Hide substance considered 
as a gel 573 

Classification of pelts . . . 573 

^reparation of the skins . . 574 

Depilation processes . . 575 

Liming 575 

Sweating 575 

Beaming 576 

Bating 576 

Tanning processes .... 577 
With tannins (vegetable 

tannage) 577 

Sole leather 578 

Upper leather .... 578 

Currying 579 

Colored leathers . . . 579 

Split leathers (skivers) . 579 

Tawing (mineral tannage) 580 

Chrome tannage . . . 580 
Combination tannage 

(dongola process) . . 580 
Tanning with oils . . .581 

Degras 581 

Sod-oil 581 

Morocco leather .... 582 

Russia leather 582 

Patent leather 582 

Parchment and vellum . . 582 

Artificial leather 583 

Theory of tanning .... 583 


Celluloid 584 

Cellulose acetate .... 585 

Bakelite . . . , 585 

Galalith 586 

Caoutchouc or India rubber . 586 
Sources of crude rubbers . 586 
Synthetic rubber .... 587 
Preparation of crude rub- 
bers for manufacturing . 587 
Preparation of rubber 
"compound" .... 588 

Vulcanizing 588 

Reclaimed, recovered, or 

devulcanized rubber . . 589 

Rubber substitutes .... 589 

Rubber cement 591 

Ebonite, hard rubber, or vul- 
canite 591 

Gutta-percha 591 






Ore dressing 593 

Wet processes 593 

Dry processes 593 


Oxidizing roast ..... 594 

Sulphatizing roast .... 594 

Chloridizing roast .... 594 

Reverberatory furnace . . 595 

Ropp furnace 596 

McDougal furnace . . . 597 

Howell-White furnace . 598 

Shaft furnace 599 

Heap roasting 599 

StaU roasting .599 

Dwight-Lloyd sintering ma- 
chine . t ; / * . . . . . 600 


Ores of iron ."V . V-. . . 601 
Blast furnace for iron . . .601 
Chemistry of the blast- 
furnace process . . . 602 
Pig iron . '. ;. . , . . 604 
Wrought iron :.../.,. . 604 
Steel . . . -, .... 605 
Bessemer process .... 605 
Acid process .... 606 
Basic process .... 607 
Open hearth process . . 607 
Campbell furnace . . . 608 
Monell process . . . 609 
Crucible process .... 609 
Cementation process . . 609 

Special steels 610 

Electrical methods for steel 
making 610 


Ores of copper 611 

Reverberatory smelting . 611 
Blast-furnace smelting . . 613 
Comparison of rever- 
beratory and blast- 
furnace .615 

Copper converting . . . 615 
Leaching processes for cop- 
per 616 

Longmaid process . . 617 
Copper refining . . . .617 
Properties and uses of 

copper . . . . . . 618 


Ores of lead 618 

Blast-furnace smelting of , 
lead 619 

Reverberatory smelting of 
lead 620 

Ore hearth for lead smelt- 
ing 620 

Refining of lead . . . .620 
Parkes' process . . .621 
Pattinson's process . . 622 
Cupellation . . . . .622 

Properties and uses of lead . 622 


Ores of zinc 623 

Reduction of zinc in clay 

retorts 623 

Refining of crude zinc . . 624 
Properties and uses of zinc . 625 


Occurrence and extraction 

from its ores . . ' . . 

Properties and uses of 

cadmium . . 




Occurrence of tin ore . 
Smelting of tin . . 
Refining of crude tin 


Ores of silver 628 

Direct extraction of silver 

from its ore .... 628 
Cyanide process . . . 628 
Amalgamation . . . 628 
Patio process .... 628 



Washoe process 
Reese River process 
Leaching processes 




Ores of gold 630 

Extraction of gold from its 

ores 630 

Placer working . . . 630 
Amalgamation .... 630 
Cyanide process . . .631 
Precipitation of gold 
from cyanide solu- 
tion with zinc . . 632 
Siemens-Halske elec- 
trical method of 
precipitation . . . 633 
Betty-Carter process 

of precipitation . . 633 
Chlorination process of 

extraction .... 633 
Parting of gold and silver 

by use of acids . . . 634 
Miller process of parting 635 
Wohlwill electrical 

method of parting . . 635 
Moebius electrical pro- 

. 636 


Occurrence and ores of 

platinum 636 

Extraction and refining . . 636 
Properties and uses of 
platinum 637 


Ore and extraction of mer- 
cury 637 


Production by use of the 
electric furnace .... 638 

Bauxite as a source of alu- 
minum 639 

Properties and uses of alu- 
minum 639 

Alloys of aluminum .... 639 


Ores of nickel 640 

Extraction of nickel from 

its ores 640 

Orford process .... 640 
Mond process .... 640 
Browne electrolytic pro- 
cess 641 

Blast-furnace smelting 

of garnierite .... 642 
Properties and uses of 
nickel 642 


Production by electrolysis of 
fused caustic soda . . . 643 


Occurrence and extraction 
from its ores 643 


Occurrence and extraction . 644 


Occurrence and ores of bis- 
muth . . 645 

Extraction and refining of 

the metal 645 

Properties and uses of bis- 
muth 645 


Production by electrolysis 
from carnallite .... 646 

Properties and uses of mag- 
nesium 646 

Magnalium 646 


Properties and constitution 

of alloys 646 

Preparation of alloys . . 647 

Brass 647 

Bronze 647 

Bearing metal or Babbitt 

metal 647 

Solders 648 

Type metal 648 

Fusible alloys . . . .648 
Coins . . 648 


Chimie Industrielle. A. Payen. Paris, 1867. 

Grundriss der chemischeri Technologie. H. Post. 

Abriss der chemischen Technologie. C. Heinzerling. Berlin, 1888. (T. Fischer.) 

Trait^ de Chimie applique"e a 1'Industrie. Adolphe Renard. Paris, 1890. 

Handbuch der chemischen Technologie. Dr. 0. Dammer. 5 Vols. Vol. I, 

1895. Vol. II, 1895. Vol. Ill, 1896. Stuttgart. (F. Enke.) 
Chemical Technology. R.Wagner. Translated by Wm. Crookes. New York, 

1897. (D. Appleton and Co.) 
Encyclopaedisches Handbuch der technischen Chemie. F. Stohmann und 

Bruno Kerl. Vol. I, 1888. Vol. II, 1889. Vol. Ill, 1891. Vol. IV, 1893. 

Vol. V, 1896. Vol. VI, 1898. Vol. VII, 1900. Braunschweig. (F. 

Chemical Technology. Edited by C. E. Groves and William Thorp. 

Vol. I, Fuel, 1889. Vol. II, Lighting, 1895. Vol. Ill, Gas Lighting, 1900. 

Vol. IV, Electric Lighting, 1903. 
Handbuch der chemischen Technologie. Dr. Ferdinand Fischer. 2 Vols. 

Vol. I, 1900. Vol. II, 1902. Leipzig. (O. Wigarid.) 

Lehrbuch der chemischen Technologie. Dr. Ferdinand Fischer. Leipzig, 1903. 
Handbook of Chemical Engineering. G. E. Davis. 2d ed. 2 Vols. Man- 
chester, 1905. 

Trait6 Chimie Applique"e. C. Chabrie. 2 Vols. Paris, 1905. (Masson et Cie.) 
Chemische Technologie der Neuzeit. Edited by O. Dammer. 3 Vols. Stuttgart, 

1910. (Enke.) 
Chemistry for Engineers and Manufacturers. Bertram Blount and A. G. 

Bloxam. 2 Vols. London, 1905. (Griffin and Co.) 
Modern Industrial Chemistry. H. Bluecher. Translated by J. P. Millington. 

New York, 1911. (Stechert & Co.) 
Handbook of Industrial Organic Chemistry. S. P. Sadtler. Philadelphia. 

4th ed. 1912. (J. B. Lippincott.) 
A Dictionary of Applied Chemistry. T. E. Thorpe. 2d ed. 5 Vols. London, 

1912-13. (Longmans, Green and Co.) 
Vorlesungen iiber Chemische Technologie. Dr. H. Wichelhaus. 3 tc Auf. 

Dresden, 1912. (Steinkopff . ) 
Lehrbuch der Chemischen Technologie und Metallurgie. Edit, by B. Neumann. 

Leipzig, 1912. (S. Hirzel.) 
General and Industrial Chemistry. E. Molinari. 2 Vols. Translated by E. 

Feilmann. Philadelphia, 1912. (Blakiston's Sons Co.) 

Lehrbuch der technischen Chemie. H. Ost. 8th ed. Leipzig, 1914. (Janecke.) 
Industrial Chemistry. Chapters by Specialists. Edited by A. Rogers and A. B. 

Aubert. New York. 2d ed. 1915. ( Van Nostrand Co. ) 




A. or Ann. = Annalen der Chemie und Pharmacie, by Liebig and others, 1832 -f . 

Ann. chim. phys. = Annales de Chimie et de Physique. Paris, 7 series, 1789 + . 

Ber. = Berichte der deutschen chemischen Gesellschaf t. Berlin, 1868 + . 

Bull. Soc. Chim. = Bulletin des Seances de la Socie'te' chimique de Paris, 
2 series, 1864 + . 

Chem. Centralb. = Cheinisches Centralblatt. 4 series, 1829 +. 

Chem. Ind. = Zeitschrift fur die chemische Industrie. 1878 + . 

C. N. or Chem. N. = Chemical News. 1860 + . 

C. R. or Compt. rend. = Coinptes-rendus hebdornadaires des Seances de 1' Acade- 
mic des Sciences. Paris, 1835 +. 

Chem. Zeit. = Chemiker-Zeitung. 1877 +. 

Dingl. J. = Dingler's poly technisches Journal. 1820 + . 

Electrochem. Ind. = Electrochemical Industry. 1902 -f 1909. 

Eng. Min. Jour. = Engineering and Mining Journal. 1866 +. 

Jahresb. = Jahresbericht tiber die Fortschritt der Chemie, u. s. w. 

J. Am. Chem. Soc. = Journal of the American Chemical Society. New York, 
1879 +. 

J. Chem. Soc. = Journal of the Chemical Society of London. 1849 + . 

J. Ind. Eng. Chem. = Journal of Industrial and Engineering Chemistry. 1909 -f. 

J. Soc. Chem. Ind. = Journal of the Society of Chemical Industry. London, 
1882 +. 

Met. Chem. Eng. = Metallurgical and Chemical Engineering. 1909 +. 

Trans. Am. Inst. Chem. Eng. = Transactions of the American Institute of 
Chemical Engineers. 1908 + . 

Trans. Am. Inst. Elect. Eng. = Transactions of the American Institute of Elec- 
trical Engineers. 1884 + 

Trans. Am. Inst. Min. Eng. = Transactions of the American Institute of Mining 
Engineers. 1871 +. 

Trans. Am. Electrochem. Soc. = Transactions of the American Electrochemical 
Society. 1902 +. 

W. J. = Wagner's Jahresbericht der chemischen Technologie. 1855 +. 

Zeitschr. angew. Chem. = Zeitschrift fur angewandte Chemie. Berlin, 1887 +. 

Zeitschr. anorg. Chem. = Zeitschrift fur anorganische Chemie. 1892 +. 

Zeitschr. Chem. Ind. = Zeitschrift fur die chemische Industrie. 1887 +. 

Zeitschr. Elektrochem. = Zeitschrift fur Elektrochemie. 1894 +. 

Zeit. physikal. Chem. =Zeitschrift fur physikalische Chemie. 1887 +. 



1 linear inch 



1 linear foot 

= .3048 


1 linear yard 

= .914 


1 linear mile 

= 1609. 



= 30.48 centimeters. 
= 91.44 centimeters. 
= 1.609 kilometers. 

1 cubic inch = 16.387 cubic centimeters. 

1 cubic foot = 7.48 gallons = 28.315. 

1 cubic foot of water at 16.5 C. weighs 62.355 pounds. 

1 fluid ounce = 29.574 cubic centimeters. 

1 quart = 946.6 cubic centimeters. 

1 gallon U.S. = 231. cubic inches = 3.7854 liters. 

1 gallon, U.S., of water at 16.5 C., weighs 8.3356 pounds. 

1 grain 

1 ounce Avd. 

1 pound Avd. 

1 ounce Apoth. = 

1 centimeter = 

1 meter = 

1 kilometer = 

1 liter 

1 hektoliter = 

1 gram = 

1 kilogram = 

1 cubic centimeter = 

In solutions, 

1 grain per gallon = 

1 grain per gallon x 

1 gram per liter = 

1 gram per liter = 

.064799 gram. 
28.3495 grams. 
7000. grains = 453.593 grams. 
31. 103 grams. 





1.057 quarts = 61.023 cubic inches. 

26.425 gallons. 

15.432 grains. 

2.2046 pounds Avd. = 35.274 ounces. 


fluid ounce = .272 dram. 

.017118 gram per liter. 
17.1 = parts per million. 

.008345 pound per gallon = 68.42 grains per gallon. 
.06242 pound per cubic foot. 





INDUSTRIAL chemistry deals with the preparation of products from 
raw materials, through the agency of chemical change. But there is 
an occasional exception to this definition ; for a few industries, de- 
pending on strictly mechanical changes, are classed among the chemical 
industries. Since a sharp line cannot be drawn between chemical 
and mechanical technology, a study of the former necessarily involves 
some consideration of the mechanical appliances and apparatus, by 
means of which the chemical reactions are carried ou.t. 

The products of chemical industry are exceedingly numerous and 
varied in character, but comparatively few come into the hands of 
the mass of the people for direct consumption. Many of them are 
used only in making other substances, for it is often the case that 
the finished product, by-product, or waste from one industry be- 
comes the raw material for another, and it rarely happens that one 
manufacturer, starting with the raw materials found in nature, pro- 
duces from them articles for popular use. Thus the chemical industries 
become a network of interlacing processes, and in considering one 
it is often difficult to separate it from others which have a more or 
less direct bearing upon it. Furthermore, as competition has become 
very close in many lines, the use which may be made of by-products 
and waste is so important, that processes are often carried out with the 
view of obtaining larger yields or better quality of the by-products, 


which may have become a source of considerable profit. In a few 
instances, it might be said that what were originally the by-products 
are now the chief products and main support of these particular 
industries. This is especially true in the case of the Leblanc Soda 
Industry, which would long since have been abandoned were it not 
for its production of hydrochloric acid. The utilization of waste 
materials furnishes an almost inexhaustible subject for investiga- 
tion by the industrial chemist. 

The manipulations of most frequent occurrence in the various 
processes are here defined and explained for the sake of brevity in 
the text. 


Lixiviation is the process of separating soluble from insoluble 
substances by dissolving the former in water or some other solvent. 
The mixture of substances is put into a suitable vessel, the solvent 
poured over it, and the whole allowed to stand until a strong solu- 
tion is obtained, which is then drawn off from the residue. This 
process is repeated as often as necessary, until the desired amount 
of soluble matter has been removed. Sometimes the mixture is put 
into baskets, or on gratings, which are suspended in tanks of water. 
The solution being denser than the solvent sinks to the bottom as it 
forms, and water comparatively free from dissolved material is thus 
constantly brought into contact with the substance to be lixiviated. 
The insoluble substance remains on the grating or in the baskets. 
When desired, the soluble material may be recovered from the solu- 
tion by evaporation or precipitation. Extraction is the term usually 
employed when some solvent other than water is used in lixiviating. 
Thus we speak of extraction by steam, alcohol, carbon disulphide, etc. 


Levigation is the process of grinding an insoluble substance to a 
fine powder, while wet. The material is introduced into the mill 
together with water, in which the powdered substance remains sus- 
pended, and flows from the mill as a turbid liquid or thin paste, 
according to the amount of water employed. There is no loss of 
material as dust, nor injury or annoyance to the workmen. Further, 
any soluble impurities in the substance are dissolved, and the prod- 
uct thereby purified. The greatest advantage of this process is the 
facility it affords for the subsequent separation of the product into 


various grades of fineness, because of the slower subsidence of the finer 
particles from suspension. The turbid liquid flows into the first of 
a series of tanks, and is allowed to stand for a certain time. The 
coarsest and heaviest particles quickly subside, leaving the finer 
material suspended in the water, which is drawn from above the 
sediment into the next tank. The liquid is passed from tank to 
tank, remaining in each longer than it remained in the preceding, 
since the finer and lighter the particles, the more time is necessary 
for their deposition. In some cases a dozen or more tanks may be 
used, and the process then becomes exceedingly slow, as very fine 
slimes or muds may require several weeks for the final settling. But 
as a rule, from three to five days is sufficient. 

The term " levigation " is now often applied to mere sedimenta- 
tion, a substance being simply stirred up in water, without previous 
wet-grinding, in order to separate the finer from the coarser parti- 
cles, as above. 


Evaporation, in a technical sense, denotes the conversion of a 
liquid into a vapor for the purpose of separating it from another 
liquid of higher boiling point, or from a solid which is dissolved in 
it. In the great majority of cases, the liquid evaporated is water. 
If the liquid evaporated is to be recovered, the vapors are condensed, 
and the process then becomes one of Distillation (see p. 9). 

There are four general methods of evaporation : 

1. Spontaneous evaporation in the open air. 

2. Evaporation by application of heat directly from a fire to 
the vessel containing the liquid. 

3. Evaporation by indirect application of heat from the fire, as ^ 
by means of steam, with or without pressure. 

4. Evaporation under reduced pressure. 

The first method, by spontaneous evaporation in the open air, is 
comparatively slow, and requires exposure of very large surfaces of 
liquid. The time necessary depends Upon the temperature and 
humidity of the air, and the completeness with which the vapors 
are removed from the surface of the liquid; hot, dry weather, es- 
pecially if a brisk wind is blowing, evaporates water quite rapidly. 
This process is only used for the manufacture of salt from sea water, 
or from natural brines. In certain warm countries considerable 
quantities of salt are thus prepared, and in this country some is made 
from a brine found near Syracuse, N.Y. Sometimes weak brines 

FIG. 1. 


are allowed to trickle in fine streams over tall piles or " ricks " of 
brushwood in the open air. The liquid being so exposed in thin 
layers, to the air and wind, is concentrated to such a degree that it 
will pay to complete the evaporation by artificial heat. 

The second method,* by direct application of heat from a fire, is 
very largely used in the arts. This may be done in two general 
ways : 

(a) The flames, or hot gases from the fire, are generally allowed 
to play directly on the bottom of the vessel containing the liquid; 

or they may pass through 
flues or pipes, set into the 
vessel, so that the liquid 
surrounds them on all sides 
(Fig. 1). Such pans are 
often several yards in 
length, and may contain 

one large flue, or several small ones, according to the work desired; 

but this form of apparatus is expensive to build, and difficult to keep 

in repair. 

(b) The flames and hot gases may be conducted over the surface 
of the liquid to be evaporated. This mode is only used for coarse 
and common products, or in the concentration or recovery of waste 
materials. But it has the advantage that the bottom of the pan is 
less liable to be injured by the crusting of a precipitate upon it. 
Another point often in favor of surface heating is that the liquid is 
evaporated in a reducing atmosphere. But as flue dust and ashes 
are liable to fall into the pans, the product is usually impure. Large 
shallow pans are used, which are generally arched over with brick, 
in order that the heat may be better utilized, through radiation from 
the brick walls. There are 

various ways of setting the 
pans for this process ; a simple 
method is shown in Fig. 2. A 
modification of this method is 
the use of a long cylinder, set 
at a slight incline, and revolv- 
ing about its longitudinal axis (Fig. 3). The lower end is open for 
the entrance of the flames and gases from the grate (A), which pass 
through the cylinder (B), on their way to the chimney (D). The hot 

FIG. -2. 

* To save expense, the waste heat from calcination or furn-acing operations is 
frequently utilized. 


gases are often passed through the flues of a boiler (C), to utilize the 
waste heat. The solution to be evaporated is fed into the cylinder 
at the upper end in a small stream, and comes in direct contact with 
the flame. The water is evaporated, and the solid matter is deliv- 
ered into the pit or wagon (E) at the lower end of the furnace, in a 

FIG. 3. 

dry and calcined state. Such furnaces are frequently used for evapo- 
rating waste liquors to recover the salts which they contain ; and for 
the treatment of sewage and other liquid refuse. 

The third method of evaporation, by the use of steam heat, is 
very often employed where there is danger of injury to the product 
by overheating. 

(a) Jacketed pans or kettles may be used. These are simply 
double-walled vessels, the steam being admitted between the walls. 

(6) The steam may be allowed to circulate through coils of pipe, 
placed inside the vessel, which is sometimes made of wood. The 
temperature of the liquid depends on the steam pressure ; very often 
exhaust steam is employed. 

The fourth method, evaporation in vacua, is merely a modifica- 
tion of either the second or third method, but is considered sepa- 
rately for convenience. The boiling point of a liquid may be very 
materially lowered by reducing the pressure within the vessel. Hence, 
solutions containing substances which would be injured by the heat 
necessary to boil them under the atmospheric pressure, or liquids 
boiling at very high temperatures, are evaporated in vacuum pans. 

The different forms of apparatus used for vacuum evaporation 
vary much in their details, but all depend on the principle of re- 
duced pressure. The essential parts of the plant are the vacuum 
pan or still, the pump for exhausting the air and steam from the 
pan and sending them to the condenser, and the heating apparatus. 
The vacuum pan is usually a globular copper or iron vessel, pro- 
vided with a manhole, a pressure gauge, and a discharging valve. 
Very often a piece of heavy plate glass is set in the side to afford a 


view of the interior during evaporation. On the top of the pan is 
a dome or short tower, from which a pipe leads to a receptacle, called 
the " catch-all," that retains any liquid which may escape from the 
pan. A small pipe returns this liquid to the pan, and a larger one 
connects the " catch-all " with the vacuum pump, which is an ordi- 
nary double-cylinder air pump of large size, driven by an engine. 
An injector pump, which condenses the steam directly, may be used. 
The pan is generally heated by steam coils within it, or by a steam 
jacket, or by both. 

A very efficient method of vacuum evaporation is that obtained 
by the use of Multiple Effect Systems. In these greater economy of 
fuel for heating is secured. The apparatus consists usually of two 
or more simple vacuum pans, so joined together that the steam from 
the boiling liquid in the first pan is made to pass through the coils 
and jacket of the second pan, and the steam generated in the second 
pan goes through the coils and jacket of the third, and so on through 
the system. The vacuum maintained in each pan of the series is 
greater than in the one preceding. Hence, notwithstanding its 
increased concentration, the boiling point of the Uquid in the second 
pan is so low, that the steam from the first pan is sufficiently hot to 
boil it. Similarly the steam from the second pan is made to boil 
the liquid in the third, in which there is still less pressure, and so on 
to the last pan, in which the highest vacuum is maintained. As a 
rule only four pans are used, for it is very difficult to sustain the 
vacuum sufficiently to work another pan in the series. In many plants 
only three pans (triple effects) are used. 

An effective modification of this method is the apparatus known 
as the Yaryan evaporator (Fig. 4). It is made in triple and quad- 
ruple effects, and each pan is exactly like its neighbors. It consists 
of an outside shell of iron, within which is a system of small tubes 
(A, A), joined together in groups of five or six, each group constitut- 
ing a section or unit. The tubes in each unit are so connected at 
the ends as to form one continuous coil. The liquor to be evaporated 
is run through the several coils thus constructed in each pan. The 
tubes in the first pan are heated by steam, introduced into the shell 
directly from a boiler. As the liquid flows through the tubes, it is 
brought to boiling, and the steam generated mingles with it, convert- 
ing the whole mass into foam, which runs through the coil and spurts 
against a baffle plate in the " separator " (B, B), which is an enlarged 
chamber at the end of the shell. The steam and liquid are separated, 
the liquid falling to the bottom and running off into the receiver (C), 


to be passed through the tubes of the next pan. The steam rises, 
passing through the steam dome and "catch-all " (D), and into the 
shell of the next " effect," through the coils of which the liquid is 
passing under still greater vacuum, and so on through the system. 
The apparatus is very economical in its use of fuel, and as the liquid 
is exposed in thin films to the heat, the evaporation is rapid; hence 
the liquid is subjected to a high temperature for only a short time. 

FIG. 4. 

The apparatus is nearly automatic in its action, and needs little atten- 
tion. It can be stopped and started quickly, since it contains only 
a small quantity of liquid at one time, and it occupies but little floor 
space when the several " effects " are placed one over the other. 

The ordinary form of vacuum pan evaporates about 8j Ibs. of 
water per pound of coal, but it is said that the best forms of Yaryan 
apparatus evaporate from 23 J to 25 Ibs. of water per pound of coal 
in a triple effect, and 30J Ibs. in a quadruple effect.* 

The Kestner evaporator utilizes the " climbing film " principle, 
by which the liquid to be evaporated is automatically distributed 
over the heating surface, without the use of pumps. The apparatus 
is built in multiple effect and each pan (Fig. 5) consists of a narrow 
vertical shell (M) containing a number of small tubes (R) through 
which the liquor passes upward from the bottom. The tubes are 
tightly fixed into plates at top and bottom and are entirely surrounded 
and heated by steam in the shell (M). The tubes are about twenty- 
three feet long and are open at top and bottom. Liquor is fed in 
through the valve (V) and the supply so limited that a relatively 

* J. Soc. Chem. Ind., 1895, 112. 



small quantity enters each tube, where it at once begins to boil: 

the vapor evolved rushes up the tube, carrying some of the liquor 
along and distributing it in a thin film on the 
hot tube wall. Emerging from the top of the 
tubes the foaming mixture of vapor and liquor 
is discharged against the vanes of a centrifugal 
separator (D) by which the concentrated liquor 
is whirled against the walls of the enlarged 
vapor space (S). The liquid flows down the 
walls and passes out at (L), while the vapor 
rises through (B) and passes to the shell of the 
next effect, or to the condenser. The separa- 
tion of vapor and liquor is claimed to be so 
complete, that practically no entrainment re- 
sults : only a small quantity of liquid is in the 
apparatus and the time of heating is short; 
only about two minutes being required for the 
passage through the tubes. The drop in tem- 
perature between any two pans of the series is 
small, ranging from 8 to 12 C. Thus four and 
five effects are often used in series. The appa- 
ratus occupies little floor space, gives little 
trouble from scaling of the tubes, and is easily 
washed out. 

The Lillie evaporator is a very efficient type 
of multiple effect (Fig. 6). Slightly inclined 

straight tubes (A) tightly fastened at 

one end in the thick plate (C) open 

into the steam space (B). The other 

ends of the tubes are closed, except 

for a small air vent, and are unsup- 
ported. Thus they expand and con- 
tract freely, preventing strains and 

resulting leaks. In the upper part of 

the effect is a row of distributing 

pipes (D), each having a longitudinal 

slot on its upper side. These pipes 

are closed at one end ; the other 

opens into a distributing box (E). 

The liquor to be evaporated enters 

through (G), passes into (D), and FIG. 6. 

FIG. 5. 


flowing from the slots in thin films, is showered uniformly over the hot 
tubes (A), from whose outer surface the evaporation takes place. The 
liquor drips from tube to tube, collecting in the float box (F), from 
which the suction pipe of the centrifugal pump (H) draws it, to again 
pass over the tubes. The float in the box (F) operates a valve which 
allows fresh liquor to enter the effect just fast enough to replace that 
vaporized and what passes from the discharge pipe (J) as concentrated 
liquor. On (J) is a regulating valve governing the level of the liquor 
in (F) and thus controlling the rate of feed ; the slower the discharge, 
the greater the concentration. The float completely closes the feed 
valve when the liquor rises to a definite height in (F) ; the discharge 
valve in the last effect thus automatically controls the flow of the 
liquor from effect to effect, by influencing the action of the feed 
valves. The tubes (A) are heated by live or exhaust steam, or by 
vapor from the preceding effect, which enters the steam chamber (B) ; 
the hot water condensed in (A) collects in the bottom of (B), and pass- 
ing the steam trap goes to the steam space of the next effect; thus 
being under great vacuum, it gives up part of its heat as steam, which 
assists in the heating of this effect. The vapor from each effect also 
enters the steam space (B) of the next. 


Distillation is the process of vaporizing a liquid and recovering it 
by condensing the vapors. The liquid formed by this condensation 
is called the distillate. Distillation is chiefly employed to separate 
a liquid from non- volatile matter dissolved or suspended in it ; or to 
separate one liquid from a mixture of liquids of different boiling 
points; that one having the lowest boiling point being the first to 
begin to pass off as vapor. 

The separation of two miscible liquids by distillation depends on 
the difference between the composition of the vapor and of the boil- 
ing liquid from which it comes ; * and while never perfect is more 
complete the greater the difference in composition. Most liquid 
mixtures evolve a vapor containing more of the low-boiling constituent 
than does the liquid itself; but in some cases the reverse is true be- 
tween certain limits of composition, and such liquids always give 

* The vapor pressures of the pure liquids, or what is practically the same thing, 
their boiling points, are not the essential factors ; thus both glycerine and water, 
and hydrochloric acid and water, differ widely in vapor pressures and boiling points, 
but glycerine and water are easily separated by distillation while the separation of 
hydrochloric acid and water by this means is impossible. 


mixtures of maximum or minimum boiling point, which cannot be 
separated by distillation as the vapor and liquid compositions are in 
these cases the same.* 

During a distillation the boiling point gradually rises, and at the 
end there remains in the still a relatively small amount of the high- 
boiling liquid very free from the other component, or else a mixture 
of maximum boiling point. While the distillate has been enriched 
in the low-boiling constituent, it is far from pure, but repetition of 
the distillation improves the separation. In general it is easier to 
secure the high-boiling liquid free from the other than the reverse. 

If a mixed vapor be slowly cooled, the liquid to condense is that in 
equilibrium with the vapor; thus it is largely the less volatile com- 
ponent which separates first. By abstracting only enough heat to 
condense a part of the vapor, the remainder is greatly enriched in the 
volatile constituent ; this fractional condensation is attained by using 
a condenser with relatively hot cooling medium, the uncondensed 
vapors passing to a cold condenser for complete condensation. From 
the fractional or partial condenser, the condensate returns to the still 
for reboiling, to remove the remainder of the volatile component. 
Fractional condensation is equivalent to a redistillation, without the 
consumption of additional heat. 

The chief parts of every distilling apparatus are the boiler or still 
and the condenser. The still is usually iron or copper, and may be 
heated directly by a furnace, or by a steam-jacket, or a coil. The 
condenser is a coil of pipe, or a system of tubes, or a double-walled 
chamber, submerged in a tank of cold water. Condensers are usually 
made of iron or copper, but lead, silver, earthenware, or glass tubes 
are sometimes used. The fractional condensation apparatus is 
placed between the still and the final condenser ; it may consist of a 
series of chambers, or of pipes or U-tubes, surrounded by a water- 
bath or other liquid at a temperature between the boiling points of 
the liquids to be distilled. 

The condensate from a fractional condenser is richer in the volatile 
component than is the liquid in the still ; it is, therefore, not in equi- 
librium with the vapor from the still, and if brought into contact 

* Thus water and alcohol have a mixture of minimum boiling point at 96 
per cent alcohol, a mixture more volatile than either component ; water and 
hydrochloric acid containing 28 per cent hydrochloric acid have a maximum boiling 
point ; mixtures containing more acid than this evolve vapors richer in hydrochloric 
acid ; if less than 28 per cent acid is present, more water in proportion is given off ; 
in either case there is finally left in the still this constant boiling mixture which 
cannot be separated by further boiling. 


with that vapor, an interchange of heat results, accompanied by a 
volatilization of the low-boiling constituent and a condensation of the 
high-boiling one, which further enriches the vapor and improves the 
separation, without any additional expenditure of heat. This process 
is called dephlegmation, and to secure the maximum efficiency, must 
be repeated a number of times. The mixed vapors from the still 
are bubbled through a series of shallow layers of liquid, through which 
passes in counter-current flow, the partial condensate from the frac- 
tional condenser; this condensate, with its low-boiling constituent 
removed as completely as possible, finally flows back into the still. 
The apparatus for this, called the dephlegmator, or " column," usually 
consists of a tall tower (Fig. 8, B, B) set above the still, and divided 
by a number of perforated horizontal plates into many shallow cham- 
bers, through each of which the returning liquid must flow on its 
way to the still. The vapors from the still pass up through the open- 
ings in the plates, bubbling through the shallow layers of the liquid 
and give up heat to it, causing vaporization of its volatile constituents. 
The boiling of the liquid in all of the chambers is done by the heat 
in the vapors entering the lowest chamber : any condensation in the 
tower results in greater coal consumption to restore the heat thus 
lost. The chief object of the dephlegmator is to boil the liquid in 
the several chambers, and deliver the enriched mixed vapors to the 
fractional condenser. As radiation from the walls of the column 
causes condensation and loss of heat, the column should be covered 
or lagged. 

In an ordinary distillation, even with use of fractional condenser 
and dephlegmating column, it is only at the end of the operation that 
there remains in the still the high-boiling component in a practically 
pure state ; at any earlier stage in the process, however, it is possible 
to dephlegmate the low-boiling component from the liquid in the still, 
by using a dephlegmating column below the still. This is done in 
practice by admitting the liquid mixture to be distilled into the middle 
of a long column, the overflow from which passes down to the still 
in which there is only pure high-boiling liquid; the rising vapors 
carry the low-boiling constituent. The original mixture can now be 
admitted and the distillate and high-boiling residue withdrawn 
continuously from the apparatus. Also, the column, below the point 
of admission of the liquid to be distilled, accomplishes the redistil- 
lation of the overflow from the apparatus above, without further 
consumption of heat; in addition to the great advantage of continuity 
of operation, this type of column gives the best heat economy of any 



method of distillation, and is generally employed except where special 
conditions render it inadvisable. The heat economy can be still 
further improved if the feed liquor be heated by using it as cooling 
medium in the fractional or final condensers. 

In Coupler's still (Fig. 7) a tower (A) is placed on top of the boiler 
(B) ; between the tower and the condenser is a series of chambers 
(C, C) surrounded by a water bath, which may be kept at any 

desired temperature. While 
the mixed vapors are passing 
through the chambers, the 
high-boiling constituents are 
condensed, and the vapor 
of the more volatile liquid 
passes through (E) to the 
condenser (F). A pipe (D) 
returns the condensed heavy 
liquid to the tower, to be 
redistilled or dephlegmated. 
The French column ap- 
paratus (Fig. 8) has a series of U-tubes (C) surrounded by a water bath. 
The column or dephlegmator (B) is divided into chambers by plates, 
each of which has a central opening covered by a dome ; a small over- 
flow pipe passes from each plate to the 
next. The vapors from the boiler (A) 
pass up through the central openings 
and bubble out under the edges of the 
domes through the layer of liquid on 
each plate. The liquid thus condensed 
flows down through the overflow pipes, 
and returns to the boiler. 

The Coffey still (Fig. 9) is much 
used for alcohol and gas liquor dis- 
tillation. This consists of two towers, 
one, called the "analyzer" (E), re- 
ceiving free steam from the boiler, 
and "the other, called the "rectifier" 
(G), containing a long coil of pipe 
(C, C), through which the liquid to be 
distilled flows on its way to the ana- 
lyzer. The analyzer is divided into a series of chambers by horizontal, 
perforated plates (A) ; from each plate an overflow pipe (F) passes 

FIG. 8. 



down and dips into a shallow cup (H) on the next plate below and 
holding liquid enough to form a hydraulic seal at the lower end of 
each overflow pipe. These pipes project about an inch or so above 
the plate in which they are set, thus determining the depth of the 
liquid layer on each plate. The rectifier is divided into chambers 
by perforated plates, but has overflow pipes in its lower half only. 
In the chambers lie the coils of pipe (C) through which the liquid to 
be distilled passes on its way to the analyzer. This still works as 

FIG. 9. 

follows : Steam from the boiler is blown through (K) into the ana- 
lyzer, and passes from the top of the analyzer through the pipe (L) 
to the rectifier. The liquid to be distilled is pumped through the 
pipe (B) and the coil (C) in the rectifier, and is delivered at the top 
of the analyzer through the pipe (D). The cold liquid is heated 
by the steam surrounding the coils, and is delivered hot into the 

Since steam is being forced up through the perforations, the liquid 
cannot pass down through them, but is forced to spread out over 
the plate, and run down the overflow pipe (F) to the next plate, and 
so through the analyzer. The steam, bubbling up through the thin 
layers of liquid, heats it very hot, and causes the volatile substances 
to distil off with the steam. This mixture of steam and volatile 
matter passes from the top of the analyzer, through (L), to the bot- 
tom of the rectifier. During its passage up the rectifier, the steam 
is condensed by coming into contact with the cold pipes (C, C), 


through which the liquid is flowing to the analyzer. Thus only the 
more volatile matters pass out at the top of the rectifier, and go to 
the condenser (O). The water condensed in the rectifier contains 
some volatile matter, so it is pumped to the top of the analyzer and 
mixes with the fresh liquor to be distilled. From the bottom of the 
analyzer a waste pipe (J) carries off the spent liquor which has been 
deprived of its volatile matter. The rectifier (G) is a combination 
of fractional condenser and dephlegmator ; the analyzer (E) is strictly 
a dephlegmating column. 

Distillation with use of a dephlegmating column is sometimes 
called rectification, but this term should be reserved for such an oper- 
ation conducted so as to eliminate one or more undesirable constit- 
uents. (See Alcohol, p. 459.) 

Distillation in vacuum is sometimes employed, and will be de- 
scribed in connection with the industries in which it is used. 


Sublimation is the process of vaporizing a solid substance and condensing 
the vapors to again form the solid directly, without passing through an intermediate 
liquid state. There are very few substances which vaporize without melting, but 
in all cases of sublimation, the change from the vapor to the solid state is direct, 
and without any formation of liquid. The sublimed body is recovered unchanged 
chemically, but its physical properties are often more or less altered. In most cases, 
the temperature does not exceed a low red heat. Dissociation often occurs in the 


Filtration is the process of separating suspended solid matter 
from a liquid, by causing the latter to pass through the pores of 
some substance, called a filter. The liquid which has passed through 
the filter is called the filtrate. The filter may be paper, cloth, cotton- 
wool, asbestos, slag- or glass-wool, unglazed earthenware, sand, or 
other porous material. 

Filtration is very frequently employed in chemical technology, 
and it often presents great difficulties. In most technical opera- 
tions, cotton cloth is the filtering material, but occasionally woollen 
or hair cloth is necessary. The cloth may be fastened on a wooden 
frame in such a way that a shallow bag is formed, into which the 
turbid liquid is poured. The filtrate, in this case, is cloudy at first, 
but soon becomes clear, and then the turbid portion is returned to 
the filter. Filtration is often retarded by the presence of fine, slimy 
precipitates, or by the formation of crystals in the interstices of 
the cloth, from the hot solution. Any attempt to hasten filtration, 


by scraping or stirring the precipitate on the cloth, will always cause 
the filtrate to run turbid. 

A better form is the " bag-filter," which is a long, narrow bag of 
twilled cotton, supported by an outside cover of coarse, strong netting, 
capable of sustaining a considerable weight and hydrostatic pressurec 
These bags are often five or six feet long, and eight inches or more in 
diameter. The open end of the bag is tied tightly around a metallic 
ring or a nipple, by which the whole is suspended, and through which 
the liquor to be filtered is introduced. When hot liquids are filtered, 
the bags are often hung in steam-heated rooms, the temperature 
being nearly that of the liquid. 

In pressure filtration, the liquid is forced through the interstices 
of the filter by direct atmospheric pressure, the air being exhausted 

FIG. 10. 

from the receiver; or by hydrostatic pressure, obtained either by 
means of a high column of the liquid, or by a force pump: By the 
first method, called suction filtration, the liquid may be forced down- 
ward through the filter into a receiver; the precipitate collects on 
the top of the filter and becomes a part of the filtering layer. This 
sometimes causes difficulty, for the particles of certain precipitates 
unite to form an impervious layer. Or the filtrate may be drawn 
upicard through the filter, which is suspended in the liquid to be 
filtered ; thus clogging does not occur so easily, as a large part of the 
precipitate settles to the bottom of the vessel and does not come in 
contact with the filter until most of the liquid has been drawn off. 

In technical work, pressure is usually obtained by the filter press 
(Figs. 10 and 11). This is a strong iron frame, in which a number 
of cells of iron or other metal are supported and tightly clamped by 



FIG. 11. 

the screw (H). Each cell is made up of two flat metal plates (A), 
with planed edges, which are separated by a hollow " distance frame " 
(B). Between the filter plates (A) and the "distance frames" (B) 
are stretched the filter cloths (C), which are held in place by the 
clamping of the edges of the plates and frames. The face of each 

plate is channelled by grooves 
leading to an outlet (D) at the 
lower edge of the plate. In a 
corner, or at one side of each plate, 
distance frame, and filter cloth is 
a hole (E) in such a position that 
when clamped in the press the 
holes form a continuous channel 
(E, E, E) through the series of cells. 
This forms the feed channel 
through which the material to be 
filtered enters the cells; a side 
opening from this channel in each 
distance frame admits the material 
to the space between the plates. 
The liquid passes through the filter cloth (C) into the grooves leading 
to the outlet (D) and escapes through the cocks (F), while the sediment 
retained by the cloth accumulates in the distance frames, forming a 
solid cake, which finally fills each cell completely. 

A powerful pump supplies a continuous stream of the liquid and 
forces the sediment into the cells, where it collects in a cake and 
offers increasing resistance to the passage of the liquid. The limit 
of pressure employed to force the liquid through depends on several 
factors, and is usually determined by experiment for each material 
to be filtered. When this pressure limit is reached, the process is 
stopped, the cell taken apart, and the cake of sediment removed; 
then the cells are returned to the press frame, clamped in position, 
and the operation repeated. The air chamber (G) equalizes the pres- 
sure during the working of the pump. 

Another type of press is the central feed machine in which the 
feed channel (E, E) passes through the middle of each plate (see Fig. 10). 
In each filter cloth, to correspond with this opening, there is a hole, 
around which a small clamping ring makes a tight joint. 

The number of cells in a single press may range from half a dozen 
to a couple of hundred, according to the amount of material to be 
filtered. The average sizes of frame are from 18 to 36 inches in di- 



ameter, and the width of the frames, which determines the thickness 
of the cake, may be J inch to 3 inches. The proper size and thickness 
of cake must be determined by experiment for each material. Very 
often the filter press is fitted with special arrangements for washing the 
cake to remove the soluble matter. This is usually accomplished by a 


FIG. 126. 

special feed channel from 
which the wash-water is forced 
through the cakes as they rest 
in the cells. Sometimes the 
cells are surrounded with 
jackets for steam heating, or 
refrigeration, for filtration at 
high or low temperatures. 

The leaf filter consists of 
a closed chamber containing 
numerous filter leaves, made 
of thin wood slats or metallic 
wire webbing, covered on both 
sides by the filter cloth; the cloth is stitched or clamped around 
the edges of the leaf. A nipple at the top or side of the leaf, con- 
necting with its interior, serves as the means of suspending the leaf 
in the chamber, and also as exit for the filtrate, which passes into the 
filtrate conduit. A shut-off valve is placed between the nipple and 
the conduit. 

In Fig. 12 * is shown a leaf filter of the " clam shell " type, in 
which the chamber consists of two semi-cylindrical castings, the 
upper one fixed in suitable supports, ahd the lower one so hinged that 
it can swing away from the top half, permitting the dumping of all 
the cakes at once, without disturbing the filter leaves. A rubber 

* Jour. Ind. Eng. Chem., 1914, (VI) 143. 



gasket renders the joint tight when the filter is closed. A short glass 
tube between the outlet nipple and the filtrate conduit makes it 
possible to detect a broken filter cloth by the turbidity of the liquid 

The chamber is filled under pressure with the liquid to be filtered 
and the leaves are thus entirely submerged ; the liquid passes through 
the cloth into the interior spaces of the filter leaf, and thence out by 
the nipple to the filtrate conduit. The solid matter is retained on the 
cloth and coats both sides of the leaves, forming cakes of uniform 
thickness on each. When the cakes are sufficiently thick, the feed 
is stopped and the cakes washed without disturbing them, by intro- 
ducing water through the sludge inlet, or by a special manifold pipe 
delivering into the spaces between the cakes. After the washing, the 
residual liquor is drained out of the chamber, and the cakes freed 
from most of the retained liquor by compressed air admitted to the 
chamber. The cakes are then loosened from the cloths by introducing 

compressed air, steam, or water 
through the filtrate outlet ; this 
passes into the space inside the 
leaf frames, and escapes through 
the cloths, forcing the cakes from 
the filtering surfaces. 

Another type of leaf filter 
has the frames suspended longi- 
tudinally in the chamber, and 
the outlets for the filtrate pass 
through the press head. The 
press head and frame carrying 
the filter leaves are supported 
on a carriage, by which they can 
be moved out of the chamber 
for discharge of the cakes. This 
press is set at an incline of 8 
to 9, to facilitate movement 
of the carriage when the leaves 
are all covered with cakes. 
The centrifugal machine (Fig. 13) is much used to separate liquid 
from solid matter. It works rapidly and leaves the substance almost 
dry. It is a cylindrical cage or basket (A) of wire gauze or perforated 
sheet metal, fixed on a vertical shaft (B) which rotates at very high 
speed. The contents of the cage are thrown against the perforated 

FIG. 13. 


wall by the centrifugal force, the solid matter being held by the screen ; 
the liquid passing through impinges upon the fixed casing (C) sur- 
rounding the rotating cage. These machines vary in size from 12 to 
60 inches diameter, and 8 to 30 inches depth of basket. Two forms 
are in use: the over-driven type, in which the driving pulley (P) 
is fixed at the upper end of the shaft, above the basket ; and the under- 
driven type, in which the cage is placed on the upper end of the 
shaft, and the pulley below. In the over-driven type, the shaft is 
usually hung in flexible bearings, so the cage may adjust itself to any 
change of the centre of gravity, caused by unequal loading, and runs 
without vibration. 

Sand filters are sometimes used for work on a large scale. These 
are made as follows : Into a box having a perforated bottom is put a 
layer of coarse gravel ; this is covered with finer pebbles ; these by 
sand, and a jute or canvas cloth covers the whole. A wooden or iron 
grating is added to protect the filter when the sediment is shovelled 
out. The filter is often placed over a receptacle from which the air 
may be exhausted, thus affording pressure filtration if necessary. 


Crystals are chemically homogeneous bodies, usually having 
regular polyhedral forms, and whose molecules have arranged them- 
selves regularly according to definite laws. The tendency to form 
crystals is common to almost all chemical compounds under certain 
conditions, the forms of the crystals being characteristic of the sub- 

Crystals may form from a fusion, or by sublimation; but crys- 
tallization almost always takes place from solution. 

In general, the solubility of a substance increases as the tempera- 
ture of the liquid rises; when the boiling point is reached, under 
atmospheric pressure, the rise in temperature ceases, and no more of 
the substance dissolves. When a liquid has dissolved all of a solid 
that it can hold in solution at a certain temperature and pressure, it 
is said to be saturated for that temperature. Any decrease in the 
temperature results in the separation of a part of the substance 
usually as crystals. There are a few instances where the maximum 
solubility is reached at temperatures much below the boiling point 
of the solution, the most notable of these salts being sodium car- 
bonate and sodium sulphate, both reaching the maximum solubility 
below 35 C, During the formation of the crystal, there is a ten- 


dency to exclude from it all matter not homogeneous with it ; hence 
this is an excellent method of purifying salts. But if a concentrated 
solution, which is very impure, is allowed to crystallize, the impuri- 
ties may become enclosed in or entangled among the crystals as 
they form, producing an impure product. This can often be pre- 
vented by stirring the solution while crystallizing, thus causing the 
formation of very fine crystals or " crystal meal," which may be 
more readily washed free from mother-liquor and impurities. The 
liquid from which the crystals have deposited, is called the mother- 
liquor; it contains the greater part of the soluble impurities present 
in the original solution, and also a considerable quantity of the salt, 
which has not deposited as crystals. The amount of the latter 
depends upon the temperature at which the crystallization took 
place. By further evaporation more crystals may be obtained, but 
they are less pure than those first separated. Thus the impurities 
accumulate in the mother-liquor, and in many cases, being valuable 
salts themselves, are recovered, and add to the profits of the indus- 
try. On the other hand, the mother-liquors from some processes 
are the cause of much annoyance and expense to the manufacturer, 
since from their corrosive, poisonous, or offensive nature, they can- 
not be run into the streams or" sewers, and their disposal in some 
other way becomes necessary. 

If a concentrated solution is allowed to stand quietly while crys- 
tallizing, especially if there is a considerable quantity of the liquid 
and the temperature falls very slowly, the crystals formed are usually 
large and well defined ; on the other hand, if it be stirred, the crystals 
are small and imperfectly developed, constituting the crystal meal 
above mentioned. Since large crystals are compact and offer a 
relatively small surface to the action of water, they dissolve very 
slowly, unless pulverized. Crystal meal dissolves more readily, and 
for this reason is becoming more and more popular with manu- 


Calcination is the process of subjecting a substance to the action 
of heat, but without fusion, for the purpose of causing some change 
in its physical or chemical constitution. The objects of calcination 
are usually : (1) to drive off water, present as absorbed moisture, 
as " water of crystallization," or as " water of constitution " ; (2) to 
drive off carbon dioxide, sulphur dioxide, or other volatile constit- 
uent ; (3) to oxidize a part or the whole of the substance. There 
are a few other purposes for which calcination is employed in spe- 



cial cases, and these will be mentioned in their proper places. 
The process is often called " roasting, " firing," or " burning," by 
the workmen. It is carried on in furnaces, retorts, or kilns, and 
the material is often raked over or stirred, during the process, to 
secure uniformity in the product. 

The furnaces used for calcining substances vary much in their 
construction, but there are three general classes: reverberatory, 
muffle, and shaft furnaces or kilns. 

FIG. 14. 

Reverberatory furnaces are built in many forms, but in all cases 
the flames and hot gases from the fire come in direct contact with 
the material to be calcined, but the fuel is separated from it. The 
simplest and most common form is shown in Fig. 14. The fire 
burns on the grate at (G), and the flames, passing over the bridge at 
(E), are deflected downward by the low sloping roof of the furnace, 
and pass directly over the surface of the charge in the bed of the 
furnace at (B), finally escaping through the throat (F) into the chim- 
ney. The charge is spread out in a thin layer on the bed (B), and 
may be either oxidized or reduced, according to the method of firing 
and the amount of air admitted. 

The revolving furnace (Figs. 3 and 47) is a very important modi- 
fication of the reverberatory furnace. This consists of a horizontal 
or slightly inclined cylinder (B) of iron or steel plates, lined with 
fire-brick or other suitable fire-resisting material, and open at each 
end. The flames from a grate (A) at one end pass through it on their 
way to the chimney (D). The cylinder is revolved about its longi- 
tudinal axis by means of a gear. It is turned until a manhole in 
the side is brought directly under a hole in the floor above, the bolted 
cover is removed, and the charge dumped in. The revolution of the 
cylinder stirs the charge thoroughly, and brings it into intimate con- 



tact with the flame. To discharge the contents, the manhole cover 
is removed, the cylinder is rotated, and the material drops out upon 
the floor or into a car placed for it. To facilitate discharging, the 
lining usually slopes from all sides towards the manhole. The speed 
varies from about two revolutions a minute to one revolution in 
five or ten minutes. These furnaces are extensively used, their ad- 
vantages being the intimate mixing and even heating of the charge, 
and the large quantities, amounting often to several tons, which can 
be worked at one time. 

' G 
t r- 



| A 



^ (^aaa^a^SSSSE 


FIQ. 15. 

Muffle furnaces (Fig. 15) are so constructed that neither the fuel 
nor the fire gases come in direct contact with the material to be cal- 
cined. A retort (A) of iron, brickwork, or fire-clay is placed over 
the fire grate (G). Flues (F, F) are built around the retort, and 
through these the hot gases from the fire pass on their way to the 
chimney (E). 

Shaft furnaces and kilns are of two general classes, periodic and 
continuous. After a charge has been calcined, the periodic furnace 
(p. 175) or kiln is allowed to cool before it is emptied and recharged. 
In the continuous variety (p. 176) this is not necessary, and the cal- 
cined substance is withdrawn and fresh material added without loss 
of time or waste of heat. The furnaces may be charged with alter- 
nate layers of fuel and material to be calcined. By this method, 
known as " burning with short flame," the material to be calcined is 
in close contact with the fuel, and is of course more or less contami- 
nated with ashes. In other forms of shaft furnaces (Fig. 74) the 
fuel is burned on a separate grate, and only the flames and hot gases 
pass into the shaft; consequently, no ashes are left in the product. 
This process is called "burning with long flame." 

Any of the various forms of furnace here mentioned may be 
heated by natural gas, generator gas, or oil. This is very advanta- 
geous in the matter of cleanliness and of regularity of temperature. 
(See Fuels.) 



Since refrigerating machines have made artificial cooling of rooms 
and of material possible, industries which were formerly only carried 
on in cold weather are now operated at all seasons. The manufacture 
of ice is also a large and increasing industry, and is apparently forcing 
the natural product from the market more and more each year. 

The principle involved in a refrigerating machine is the rapid 
absorption of heat by the rapid evaporation of a volatile liquid. 
The substances most used are liquefied ammonia, sulphur dioxide, 
carbon dioxide, and the very volatile liquids derived from petroleum, 
chiefly cymogene and rhigolene. In this country by far the greatest 
number of machines employ liquid ammonia. 

The gas is heavily compressed and then liquefied by passing it 
into a coil over which a large amount of cold water flows ; the liquid 
is then forced through a small opening into a large chamber or coil 
of pipe, from which the gas formed may be rapidly exhausted by a 
pump. The rapid expansion and conversion of the liquid to a vapor 
here absorbs much heat from the walls of the coil or chamber, whose 
temperature consequently falls considerably below the freezing point 
of pure water. In order to increase the external surface of the expan- 
sion coils, cast-iron disks are placed at frequent intervals on the pipe 
perpendicular to its line of direction. Only a comparatively small 
amount of ammonia or other volatile liquid is necessary for the con- 
tinuous working of the machine. Since the gas is returned to the 
compressor, it is only necessary to supply that lost by leakage. 

It is often customary to surround the expansion coils with a brine 
or calcium chloride solution, which is then pumped through coils 
or pipes in rooms to be cooled. For making ice, galvanized iron 
boxes are filled with water and immersed in the cold brine. 

In the system shown in the diagram (Fig. 16) a certain amount 
of oil is injected into the compressor along with each charge of am- 
monia. This insures complete emptying of the compressor at each 
stroke, lubricates the piston, prevents the gas escaping behind the 
piston, and absorbs part of the heat evolved in the cylinder by the 
compression of the gas. This oil is separated from the liquefied am- 
monia by gravity in separating tanks and returned to the compressor. 

The machines above described are called " compression ma- 
chines," because the volatile substance is compressed directly to be 
used again. Another class of refrigerating apparatus depends for 
the recovery of the volatile substance upon absorption of the vapors 



in some liquid from which they can again be set free. The cooling 
effect in this case is also produced by rapid evaporation of the lique- 
fied gas, the difference in the two classes of machines being in the 
method of recovering the vaporized liquid for use again. In the 
Carre ammonia absorption apparatus very concentrated aqua am- 
monia is heated in an iron retort or generator ; the ammonia gas is 
driven out through a condenser, chilled by running water over the 
coils; the ammonia is liquefied by its own pressure, and the liquid 
passes through an expansion valve into the refrigerator coils, which 



FIG. 16. 



may be surrounded with brine. The expanded ammonia vapor then 
passes to the absorber, containing weak ammonia water from the gen- 
erator, which has been cooled by running through coils immersed in 
water. Reabsorption of the ammonia vapors to form concentrated 
aqua ammonia takes place, and the solution is returned to the generator 
to repeat the cycle. Another style of absorption machine evaporates 
water in a vacuum apparatus, and absorbs the vapor in concentrated 
sulphuric acid. The dilute ac\d thus produced is concentrated in open 
pans by evaporating the water, and is used again. 

The absorption machines require a large quantity of cooling water, 
and are generally more complicated and expensive than the com- 
pression machines. 

A third class of machines are those depending on the sudden ex- 


pansion of highly compressed air or other gas, which does not liquefy 
at the temperature and pressure used. These machines are large 
and complicated and are not adapted to making ice, but find limited 
use for cooling and ventilating on board war vessels where any traces 
of ammonia or other vapor would be dangerous. 

Refrigeration, Cold Storage, and Ice-making, A. J. Wallis-Taylor. 
London, 1902. (Lockwood & Son.) 


By the specific gravity of a liquid is meant its relative weight com- 
pared with the weight of an equal volume of pure water at a definite 
temperature. This determination is one of the most frequent opera- 
tions in chemical work and may be done with a pyknometer when 
very exact results are required, but in technical operations, sufficient 
accuracy for all practical purposes may be attained by the hydrometer. 
This is usually a glass instrument, consisting of a cylindrical bulb, 
weighted at the lower end, and drawn out at the upper end to a long, 
slender tube, carrying a scale. The gradations of the scale begin at 
the top and read downward, the numerically greater reading being at 
the bottom, except in one instance, that of Baume's scale for liquids 
lighter than water. Since the specific gravity of a liquid varies as its 
temperature changes, the scale is adjusted to a certain temperature, 
usually about 15 C., at which determinations must be made. 

When the hydrometer is placed in a liquid, it sinks sufficiently 
to displace a volume of the liquid equal in weight to the weight of 
the instrument, and floats in an upright position. Should the hy- 
drometer sink so deeply into the liquid that the scale is entirely below 
the surface, the specific gravity is less than the spindle is intended 
to measure, and one having lower * numerical readings should be used. 
If, on the contrary, the spindle does not sink deep enough to bring 
the scale into the liquid, an instrument having higher numerical scale 
readings is necessary. 

Three systems of hydrometer scales are in common use, besides a 
great number of special scales intended to give one particular factor 
in the specific gravity of a liquid ; e.g. the per cent of alcohol in a 
mixture of alcohol and water, or the amount of sugar in a syrup, etc. 

The direct specific gravity hydrometer is so constructed that the 
reading on its scale shows the specific gravity of the liquid directly 
as compared with pure water at the same temperature (15 C.). Its 

* Baume's hydrometer for liquids lighter than water is an exception (p. 27). 


scale is adapted to liquids heavier or lighter than water. The point 
to which it sinks in pure water at 15 C. is marked 1.000. As usually 
furnished, a set of these hydrometers consists of four spindles, the 
scale being thus divided into four sections. The first spindle, with 
gradations from 0.700 to 1.000, is for liquids lighter than water, and 
the others are for those heavier than water. The scale is usually 
divided about as follows: 1.000 to 1.300 oh the second spindle, 1.300 
to 1.600 on the third, and 1.600 to 2.000 on the fourth. 

The gradations at the top of each spindle are farther apart than 
those at the bottom of the stem,* rendering the reading somewhat 
more difficult in dense liquids than in those of lighter gravity. 

TwaddelTs hydrometer is also a direct-reading instrument. The 
system consists of a series of spindles (usually six in number) car- 
rying gradations from to 174. The reading in pure water, at 
15.5 C., is taken as 0, and each subsequent rise of 0.005 sp. gr. is 
recorded on the scale as one additional division. Thus 10 Twaddell 
becomes 1.050 sp. gr. The gradations on this scale are also closer 
together as the specific gravity increases, but as its total length is 
divided among six spindles, the readings are not so difficult even at 
the highest gravities. The instruments are small, and may easily 
be used in an ordinary 100 cc. measuring cylinder. For the reasons 
that it is easy to read, requires but a small quantity of liquid to be 
tested, and permits a ready conversion of its readings into specific 
gravity by a very simple calculation, this is a convenient hydrom- 
eter for ordinary factory or laboratory use. It is, however, hot adapted 
to liquids lighter than water. 

Twaddell readings are converted into specific gravity as follows: 
Multiply the reading by .005 and add 1.000 to the product. Thus 15 
Twaddell becomes 1.075 sp. gr. (1.000 + [15 X .005] = 1.075.) 

Baume's hydrometer is largely used in technical work, but its 
readings bear no very direct relation to true specific gravity. Baume 
dissolved 15 parts of pure salt in 85 parts of pure water at 12.5 
C. The point to which his instrument sank in this solution was 
marked 15 ; the point to which it sank in pure water was marked 0. 
The distance between these points was divided into fifteen equal parts, 
and the entire stem marked off in divisions of this width. This 
produced an instrument for liquids heavier than water. 

For liquids lighter than water, the point to which the instrument 
sank in a 10 per cent solution of salt was marked 0, and that to 

* For the explanation of this fact consult any of the larger works on physics. 


which it sank in distilled water was marked 10 ; the distance between 
these points was divided into 10 equal parts, and this gradation con- 
tinued the entire length of the spindle. The thus being placed at 
the bottom of the stem, the lighter the gravity of the liquid tested, 
the greater numerically is the reading of the scale. For instance, a 
liquid reading 70 Be. is of less gravity than one of 50 Be., which in 
turn is lighter than water at 10 Be. 

To further complicate matters, the instrument makers have pro- 
duced instruments with erroneous scales. A test made a few years 
ago disclosed thirty-four different scales, none of which was correct ! * 

The conversion of Baume readings to specific gravity involves 
some calculation and is usually accomplished by reference to tables. 
The formulae for this conversion are as follows : 

g ^ o _ -j, r _ 145 (for liquids heavier than water. Tempera- 

sp.' gr. ture 60 F.).f 

B^Q_ 140 J^Q (for liquids lighter than water. Tempera- 
~sp. gr. ture 60 F.). V 

The pyknometer is not often used in technical work, but a 
brief description of it may not be out of place here. It consists of 
a small bottle, having ground into its neck a capillary tube enlarged 
at its upper end, to form a reservoir which is closed by a stopper. 
The tube is removed and the bottle filled with the liquid to be tested ; 
the tube is then inserted tightly, the liquid displaced rising through 
the capillary to the enlarged part of the tube. The stopper is then 
loosely inserted and the bottle placed in a bath at the temperature 
at which the gravity is to be taken. When the bottle and contents 
have reached this temperature, the stopper is taken out and the liquid 
in the reservoir removed by means of absorbent paper, until the level 
of the liquid recedes within the capillary to a mark thereon. The 
stopper is then tightly inserted and the bottle removed from the bath, 
and after cleaning and drying its outside, allowed to stand until it 
reaches the normal temperature of the room. It is then weighed, 
and the specific gravity of the liquid is calculated from its known 
volume, previously determined by calibration of the bottle. (For 
determining specific gravity by means of the pyknometer, see T. E. 
Thorpe's Dictionary of Applied Chemistry, Vol. V, pp. 107-114.) 

* C. F. Chandler, Proc. Nat. Acad. Sciences, 1881. 

t J. Am. Chem. Soc. 21 (1889), 119. J. Soc. Ind., 1905 (24), 786. 

* European instrument makers use the formula Be. = 144.3 . See Alkali- 

sp. gr. 

makers' Handbook (Lunge and Hurter), p. 175. 



Westphal's balance is used to determine the specific gravity of 
liquids. A glass plummet of known weight and volume, suspended 
from the beam by a fine platinum wire, is submerged in the 
liquid. The weight which the plummet loses by this submersion is 
the weight of the volume of liquid it displaces. The decimal gradua- 
tion of the beam, with the use of riders of 0.1, 0.01, and 0.001 part of 
the weight of the water displaced by the plummet, permits the actual 
specific gravity to be read off on the beam, as soon as the latter 
is brought to equilibrium with the plummet, suspended in the liquid in 


Particles in the interior of solids and liquids are greatly influ- 
enced by the presence in their immediate neighborhood of other 
particles ; this is indicated by describing solids and liquids as " con- 
densed phases." Particles on the surface are free from such action 
on one side, but not on the other. Hence the properties of the sur- 
face of a condensed phase are different from those of the interior; 
thus an aqueous soap solution contains more soap in its surface layer 
than in its interior, but with most inorganic salts the reverse is true. 
The surface layers are so thin, however, that these differences become 
appreciable only when the ratio of surface to volume is enormous. 
This obtains when a substance is in the form of extremely small, 
separate particles; since the voids between the particles are neces- 
sarily filled with something, such a system consists of at least two 
phases. The whole is called a disperse system the separated par- 
ticles the disperse phase, and the void-filling substance the dispersing 
medium. Surface phenomena are usually of importance only in such 
disperse systems. The following table illustrates the phenomena : 















Whipt Cream 

(Air emulsified in oil.) 




Emulsions Cream 

(Oil emulsified in water.) 




Suspensions Colloidal 

solutions, as gold in water. 








Frozen Cream Gels 




"Milk "Glass 


When the size of the dispersed particles is small, the friction be- 
tween them and the gaseous or liquid dispersing medium becomes very 
great, and relative motion is difficult. Hence separation by decan- 
tation is difficult or impossible, nitration fails except through the 
densest filtering media, and the separation of the phases is a serious 
problem. Frothing in evaporators, dust production in furnacing and 
grinding operations, the " fume " from storage batteries and pickle baths, 
are cases in point. The solution of the problem is found either in in- 
creasing relative motion by the application of an external force, 
(as in centrifugal separators for dust, cream, etc.), or in causing co- 
alescence of the particles. This may be brought about by enmeshing 
them in a mass of larger particles (as in the use of iron or aluminum 
hydroxides in water purification), by impinging upon surfaces, wet 
or otherwise (as in " catch-alls " and gas washers), or by electrical 
discharge, or by mutual co-precipitation. These last two methods 
are based on the fact that in general there exists at the surface be- 
tween two phases an electrostatic potential difference. The parti- 
cles of the dispersed phase are therefore charged with reference to the 
dispersing medium and hence are capable of electrolysis (movement 
in an electric field followed by discharge on reaching the electrode). 
Also the particles are often held apart and prevented from coalescing 
to form larger aggregates, by the mutual repulsion of the charges on 
the particles; when discharged, coagulation or precipitation results. 
Discharge may be caused* (a) by electrolysis ; (6) by neutralization 
with the charge of a disperse phase of opposite sign ; (c) by the adsorp- 
tion of ions of opposite sign. 

A characteristic property of such disperse systems is the power 
possessed by the surfaces between the phases to condense upon 
themselves large quantities of gases and of solutes from liquid phases. 
This condensation is called adsorption; the amount adsorbed x, on 

the surface m, is given by the expression = kC n , where C is the 


* An illustration of (a) is the Cottrell Process for precipitating smoke and 
dust from smelters, cement works, etc. ; the gas containing the fine solid particles 
in suspension is passed through a cylindrical metal tube or chimney, at the axis 
of which is a fine wire. Between this wire and the chimney a high direct-current 
potential difference (100,000 volts and over) is maintained, which causes the 
charged dust particles to move rapidly to the surface of the stack, discharge, and 
coalesce. Owing to the small ratio of charge to weight of dust particle, the cur- 
rent consumption (hence the cost) is small. The efficiency of dust removal and 
the gas-handling capacity are high. A technical application of (b) is in the tan- 
nage of leather (p. 577), while (c) is illustrated in the use of certain electrolytes *to 
coagulate fine precipitates (the use of salt in washing -white lead, p. 225, and the 
addition of NI^NO? to aid in filtering phosphomolybdate precipitate). 


concentration of the substance adsorbed, while k and n are constants, 
n being greater than unity, seldom less than 2 or more than 20. 
Adsorption at low concentration is relatively far greater than at high, 
and this can be used to remove small amounts of impurities from solu- 
tions, as in water purifying (p. 49), sugar refining (p. 429), decoloriz- 
ing oils (p. 342), etc. The adsorbing agent may be brought in contact 
with the liquid by admixture or by use of the counter-current prin- 
ciple, the solid being held in a suitable container through which the 
liquid passes, thus forming an adsorption filter. Adsorption is gener- 
ally greater the lower the temperature and the higher the molecular 
weight and complexity of the substance adsorbed. Certain materials, 
notably bone-char and fuller's earth, have high adsorptive power, 
while others, such as animal and vegetable fibres, are very specific in 
the substances they adsorb. 

Where the dispersing medium is a liquid, the disperse phase is called 
a colloid, if exceedingly small, so that separation by filtration is very 
difficult ; if somewhat larger, it is called colloidal, or a colloidal precipi- 
tate or suspension. Solid colloids vary much in their capacity to com- 
bine with the solvent ; the following aqueous colloids are approximately 
in the order of their degree of hydration in colloidal solutions : (1) glue, 
caseinogen, dextrins, etc. ; (2) hide substance, starch ; (3) ferric 
hydroxide, aluminium hydroxide, silicic acid ; (4) clay ; (5) arsenic 
sulphide, sulphur; (6) colloidal metals (gold, etc.). When highly 
hydrated, the colloid acquires many of the properties of a liquid, and 
the colloidal solution resembles an emulsion. Such colloids are called 
emulsion colloids or emulsoids; the less hydrated ones are called sus- 
pensoids. A sharp distinction cannot be drawn between them. 

The value of a solid colloid depends on its possession of emulsoid 
characteristics. Solid emulsoids when treated with a solvent first 
swell (glue, rubber, starch), due to combination with the solvent, 
.but finally dissolve, i.e. the individual particles separate and stay 
in suspension in the solvent. By lowering the temperature, the 
solvent in combination with the colloid increases at the expense of 
that in the dispersing medium, and if $ie latter be not too great in 
amount, it finally becomes so small relative to the colloidal phase that 
the two change functions, i.e. the system becomes a suspension of 
small drops of liquid dispersed in a solid, much swollen with solvent. 
This is called a gel or jelly, and the transformation temperature is the 
jellying point. 

Solid colloids are substances whose individual particles or mole- 
cules are very large and complex molecular aggregates. They are 


in general chemically inert and little affected by solvents ; when heated, 
or when swollen with a suitable solvent (in gel form), they are " plas- 
tic," i.e. they can be moulded, and separate pieces pressed together 
coalesce so they can be shaped at will. Some possess great strength 
and hardness and are important as materials of construction. Such 
colloids consist of one or several units U, a number, n, of which are 
associated into an aggregation U n . Thus in starch the unit is the 
dextrose radical, association resulting from elimination of water from 
dextrose ; in rubber the unit is isoprene or polyprene ; in albumens 
the units are a number of amido-acids. In general the larger the 
aggregate U n , the stronger and harder the solid colloid, the more 
inert chemically, the less soluble in solvents, and the less ductile and 
plastic the material becomes. Mechanical manipulation, heat, and 
the action of solvents tend to decrease the size of the aggregate by de- 
creasing n. 

Many important solid colloids (cellulose, hair, wool, and silk) are 
natural products, any chemical treatment of which causes disinte- 
gration of the aggregate to a greater or less extent, with results on the 
physical properties as indicated (pp. 489, 494, 499, 505, 511, etc.). 
Many others are made by first forming the colloid as a plastic and 
then destroying the plastic condition. The important industrial 
methods of doing this are: (1) producing the plastic state by means 
of heat and hardening by cooling (as in glass, p. 196, and asphalt, 
p. 347) ; (2) rendering plastic with a solvent and solidifying by evapo- 
ration of the solvent (as in artificial silk, p. 496, and sun-dried bricks) ; 
this may be succeeded by chemical changes essentially modifying the 
character of the solid (drying of varnish, p. 397 ; burning of ceramics, 
p. 212) ; (3) increasing the chemical and physical resistance and de- 
creasing the plasticity, by increasing the size of the molecular aggre- 
gate U n , by (a) co-precipitating two emulsoids to give a more com- 
plex and stable mass (as in tanning leather, p. 573) ; (6) by increasing 
the number of units, n, in each aggregate (as in the manufacture of 
bakelite, p. 585) ; (c) by increasing the complexity of the unit U by 
chemical addition (as in vulcanizing rubber, p. 588). 


Fuels are substances which, when burned with air, evolve heat 
with sufficient rapidity and in sufficient quantity to be employed 
for domestic or industrial purposes. 

There are three classes of fuel : solid, liquid, and gaseous. In the 
majority of these the essential constituent is carbon, but in many of 
them hydrogen is also an important ingredient. In rare cases sul- 
phur, phosphorus, silicon, or manganese may take part in the com- 
bustion ; but for the purposes for which fuel is ordinarily used these 
constituents are deleterious. Oxygen is sometimes regarded as 
advantageous, but not always. Nitrogen may cause a direct loss of 
calorific power, owing to its dilution of the combustible gases, but 
in most solid fuels the percentage of nitrogen is so small that its 
effect is negligible. 


The solid fuels are wood and other matter containing cellulose, 
peat, lignite or brown coal, bituminous coal, anthracite, charcoal, 
and coke. 

Wood consists of cellulose (CeHioOs)^ lignine, resins, various 
inorganic salts, and water. The quantity of water present has 
great effect on the heating value and ranges from 25 to 50 per cent 
in green wood, and from 10 to 20 per cent in air-dried wood. Wood 
cut in the spring and summer contains more water than that cut 
in the early part of the winter. A cord of hard wood, such as ash 
or maple, is about equal in heating value to one ton of bituminous 
coal ; soft woods, such as pine and poplar, have less than half this 
amount. Wood burns with a long flame and makes comparatively 
little smoke ; but its calorific power is low, averaging from 3000 to 
4000 Cal. per kilo of air-dried wood. It is, however, easily kindled, 
the fire quickly reaches its maximum intensity, and a relatively small 
quantity of ash is formed. Wood is too expensive for industrial use, 
except in a few special cases, where freedom from dirt and smoke is 

Of other cellulose materials, shavings, sawdust, and straw are 
used for fuel in some places. They are bulky and difficult to handle, 
while their heat value, which depends on the amount of moisture 
they contain, is seldom more than from one-third to one-half that 



of good coal. Such waste matter as spent tan-bark and bagasse 
(crushed sugar cane), and the pulp from sugar beets is sometimes 
used for fuel for evaporation or for steam, but owing to the large 
amount of moisture they contain, the heat value is very low. 

Peat is the product of slow decay of mosses, especially Sphag- 
nacece, under water. It is of little importance in this country, but 
is extensively used in parts of Europe where it is found. Since it 
contains a large amount of water and inorganic matter, its calorific 
power is not high, averaging from 4000 to 5000 Cal. per kilo. One 
pound of peat evaporates about 4.5 Ibs. of water. It is dug from 
the bogs and dried in the air, sometimes being heavily compressed to 
reduce its bulk. As thus prepared, it contains from 15 to 20 per cent 
of moisture and from 8 to 12 per cent ash. It is used considerably 
as a packing material, owing to its soft and spongy consistency. 

Lignite or brown coal is intermediary between peat and bitu- 
minous coal. It was probably formed from swamp plants which 
decomposed under water, and is geologically of more recent forma- 
tion than true coal. It is dark brown or black in color, and its text- 
ure is fibrous, earthy, or sometimes vitreous. It usually contains 
from 15 to 20 per cent of moisture, a large quantity of ash, and often 
a considerable amount of sulphur. It burns freely with a long 
flame, producing much smoke, and its calorific power varies from 4000 
to 6500 Cal. It is extensively used for heating steam boilers and 
evaporating pans, and for domestic fires. 

Bituminous coal is the most important of all fuels. There is 
a great variety in the kinds of coal classed under this name, but 
they differ chiefly in the amount of volatile matter, which ranges 
from 20 to 50 per cent. They were all formed from similar sources, 
the varieties having resulted from pressure and from exposure to 
heat. The specific gravity varies from 1.25 to 1.75. They are classi- 
fied according to their behavior when burning, as fat, caking, and 
non-caking. Fat coals usually have a dull lustre, are very rich in 
volatile matter, sometimes containing as much as 50 per cent, and 
burn with a long, smoky flame, sometimes caking in the fire. Non- 
caking coals are those which burn freely, with little smoke, and do 
not cake. The caking coals burn with a smoky flame and fuse or 
sinter together. 

The formation of coal is probably due to a slow decomposition 
of cellulose matter, under fresh water, by which marsh gas (CHO and 
carbon dioxide (CO 2 ) were eliminated. The composition of a typi- 
cal coal, as shown by the analysis of good samples, may be repre- 



sented by the symbol (C26H 20 O 2 ), and assuming this, the change of 
cellulose may be represented by the equation : 

6 (C 6 HioO 5 ) = 3 CH 4 + 7 CO 2 + 14 H 2 O + C 26 H 20 O 2 . 

Various changes were afterwards brought about by the heat and 
pressure within the earth's strata, and the character of the coals 
modified in many cases. Thus, more or less of the volatile constit- 
uents were removed, and the coal itself compressed to a very hard, 
compact mass. When this process went to the extreme, nearly the 
whole of the volatile constituents were expelled, and the resulting 
product is the hard coal known as anthracite. 

Anthracite coals are nearly pure carbon, are very hard and dense, 
have a very high lustre, and contain but little hydrogen or volatile 
matter. They burn with a slight flame, form no smoke, have no 
caking properties, and are difficult to ignite. Their specific gravity 
is high, being nearly 1.75 in good Lehigh coal. They have a calorific 
value of from 7500 to 8500 Cal. 

Between bituminous and anthracite coals are a number of semi- 
anthracites, which cannot be classed in either variety. 

Coal deteriorates considerably when stored, owing to the escape 
of some of its volatile constituents. There is a popular idea that 
wetting coal before burning increases its heating capacity ; but this 
is a fallacy, for a loss of heat results. 

The average composition o*f various coals is here tabulated for 
comparison : * 







Brown coal (Wyo.) . . 







Bituminous (111.) . . . 







Bituminous (Pa.) . . . 








Semi-bituminous (W. Va.) 








Anthracite (Pa.) 






Charcoal is made by the dry distillation of wood, at a tempera- 
ture of from 400 ,to 450 C. This is done in heaps, or in closed 
retorts. All the volatile matter is driven off, and the residue con- 
sists of carbon and the inorganic constituents of the wood. Good 
charcoal is porous, brittle, with conchoidal fracture, and retains the 
* U. S. Bureau of Mines, Bui. No. 29. 



form of the wood, but has only about three-fourths of the volume 
and usually about 20 per cent of the weight of wood. It burns with 
but slight flame, without smoke, and is easily ignited. Containing 
but little sulphur or phosphorus, it is especially useful in making 
some high grades of iron and steel. Its calorific power is about 
7100 Cal. 

In this country much of the charcoal is made by burning wood 
in " charcoal pits." The wood is heaped in a hemispherical pile 
around a central opening, and covered with earth and sod, leaving 
only a few small draught holes near 
the bottom. Then it is ignited at 
the centre and allowed to burn 
until the whole pile is on fire. A 
smouldering combustion takes place, 
largely at the expense of the oxygen 
and hydrogen of the wood fibre, 
forming water, carbon dioxide, and 
volatile hydrocarbons, which escape. 
The draught holes are then all closed 
and the pit is kept carefully covered 
until the fire smothers and the char- 
coal is cold. By carbonizing in pits 
nearly all the volatile matter is lost, 
or at best, only a part of the tar is 
saved and the yield of charcoal is 
only 20 per cent by weight of the 
wood. But if the process is carried 
on in retorts, a large amount of gas, 
pyroligneous acid, and tar is col- 
lected (see p. 301), and about 30 per cent of charcoal is obtained, 
together with nearly 40 per cent of pyroligneous acid and 4 per cent 
of tar. 

Coke is made by the destructive distillation of coal. It has a 
silvery white lustre, an open, porous structure, and a metallic ring 
when struck. It contains all the ash-forming materials of the coal, 
but nearly all volatile matter and sulphur have been eliminated. 
For metallurgical purposes it must be sufficiently strong to sustain 
the weight of the charge in the furnace without crushing. The 
calorific value is from 7600 to 8100 Cal. It burns without smoke and 
with but little flame and does not cake. It is made in kilns of two 
general types : The " bee-hive " coke oven (Fig. 17) is made of 

FIG. 17. 



brick, with a circular opening (A) at the top and a door (B) at the 
side, through which the coke is drawn. A part of the coal is burned, 
in order to carbonize the remainder. As a rule, no attempt is made 
to save the volatile products or the tar. The yield of coke amounts 
to only 60 or 65 per cent of the weight of the coal. 

Coking ovens in which the by-products are saved are much used 
abroad, and to an increasing extent in this country. There are sev- 
eral kinds, but the Otto-Hoffmann, the Simon-Carves, and the Semet- 
Solvay ovens are much used. In these, the ammonia and coal tar are 
recovered, and a coke suitable for metallurgical purposes is obtained. 
The waste gas is employed to heat the retorts. 

HUM l^llllim ! P s 3illil I 

FIG. 18. 

The Otto-Hoffmann oven is shown in Fig. 18. The retorts are 
narrow chambers (O) about 40 feet long, 5 feet high, and 22 inches 
wide, having doors at each end, and heated by vertical flues (T, T) 
in the walls. Coal is charged through (F, F) while the gases and 
tar pass off through (A, A) to the hydraulic main (V, V). The gas 
for heating enters from pipe (G), mixes with hot air from the re- 
generator (R), and burns in the flue (S) under the retorts, the flame 
passing up through the flues (T, T), and down through (T', T') to 
(S'), from which the products of combustion pass through the regen- 
erator (R') and heat it. After a time, the flow of gases is reversed, 
the producer gas enters through (G') and air through (R'), burning 
together in (S')> while the products of combustion escape through 
(R). The volatile matter given off from the coal passes through 
(V) to washers and scrubbers (see Illuminating Gas), which remove 
tar and ammonia, while the gas is stored in a holder, to be led, later, 
through (G, G'), and burned under the retorts. 

The Simon-Carves oven (Fig. 19) is also a long, narrow retort (A) 
with doors at each end, but the heating flues (F, F) are set hori- 
^ontally in the retort walls. The volatile matter escapes from the 



retort through (B), passes to the washer and scrubber, whence the 
purified gas goes to the holder, from which it is drawn as needed, 
through (G), and burned with hot air. 

The Semet-Solvay oven (Fig. 20) also has horizontal flues, but 
deeper and narrower retorts than the two just mentioned. Each 
retort has an independent set of flues which are placed in the retort 
lining and backed by a heavy brick retaining wall ; this supports the 
weight of the roof arch, and also holds the heat during the drawing 
and charging of the retort. Thus the flue walls can be made much 
thinner than in the ovens previously mentioned, and the oven works 

A - Charging Holes D - Chimney Canal 

B - Gas uptake p . Heating Flues 

C - Gas Inlet to Heating Flues R - Wall between Retorts 

FIG. 20. 

more rapidly, giving a larger yield of coke, and will coke coals which 
are low in volatile matter. The lining can easily be replaced without 
rebuilding the entire oven. The retorts are usually about 30 feet long 
by 16 inches wide, and 5J feet deep, and hold about 4| tons of coal 
at each charge. No regenerative heating is used, the heat being re- 
tained in the walls between the retorts. A number of these ovens have 
been recently introduced into this country and give excellent results. 



The most important liquid fuels are crude petroleum and various 
oily residues obtained in distilling petroleum, shale oil, and coal tar. 

Crude petroleum, especially the Texas and California oils, and 
the residuum from the manufacture of burning oils and lubricators, 
are the chief sources in this country. The residuum from Russian 
petroleum, called " astatki," is very extensively used in southern 

Crude petroleum is easily regulated so as to burn without smoke 
or soot, giving a steady heat and requiring no stoking. It is less 
bulky, and from two to two and a half times as efficient as anthracite 
coal. Its heat value is about 11,000 Cal.; and it evaporates about 15 
Ibs. of water to one pound of oil. One pound of coal-tar residue 
evaporates 13 Ibs. of water. 

Liquid fuel is coming into more general use every year, espe- 
cially where long flame and high temperature are desired. It is 
usually burned as spray, being forced into the furnace by a large 
atomizer supplied with an air blast or superheated steam. 


Gas fuel is important industrially, because of the cleanliness, 
economy of labor, and exact control of the heat it affords, and of the 
much greater efficiency obtained in the conversion of energy into 
work, or the development of power, by the use of internal-combustion 
engines. Since combustible natural gas is found only in restricted 
areas, artificial fuel gas is largely prepared ; but always from solid fuels. 
Liquid fuel is itself nearly as satisfactory as gas, and is too expensive 
as a material for gas making. 

Natural gas exists already formed in the earth, and is obtained 
by boring tube wells similar to petroleum wells. Its essential heat- 
producing constituents are methane (CH 4 ) and hydrogen. It is the 
cheapest and most efficient of all fuels, when properly burned, hav- 
ing a heat value of about 9400 Cal. per cubic metre ; but it requires a 
large amount of air for its combustion, and special burners must be 

Destructive distillation of solid fuels containing a large propor- 
tion of volatile matter is the simplest method of obtaining combus- 
tible gases. This is done with both wood and soft coal, but in the 
case of the former the expense precludes its use for gas production 


Coal gas (p. 314) is made by distilling bituminous coal in retorts 
at such high temperature that the hydrocarbon constituents break 
down, chiefly into methane and hydrogen (about 40 per cent of each), 
with small amounts of carbon monoxide, carbon dioxide, nitrogen, 
and unsaturated hydrocarbons which give luminosity to the flame. 
It is primarily made for illumination, but is often used for power and 
heating, where cheaper gas is not available. 

Gas formation is important in the burning of soft coal, since dis- 
tillation results from the heat of the fire, and the gases set free in the 
furnace burn with a " long flame," developing their heat of combus- 
tion uniformly along the entire . length of the furnace. 

Besides the gas from the volatile constituents, the carbon of coal 
may be converted into a combustible gas, carbon monoxide, by burn- 
ing the carbon with a limited air supply. A disadvantage of the pro- 
cess lies in the fact that the large heat of formation of carbon mon- 
oxide (29.0 Cal. per Mol), is developed as sensible heat in the gas 
and is lost if the gases are cooled previous to their combustion. 
But if the hot gas can be used immediately without cooling, as is 
often done in lime kilns, etc., there is much economy attained. This 
may be considered as a type of " long-flame burning " of the coal. 

Water gas contains much carbon monoxide and is made by the 
action of steam on carbon at high temperature : 

C + H 2 O = H 2 + CO - 29.0 Cal. 

There is, however, a very large absorption of heat (29.0 Cal.) in this 
operation. But the gas has high calorific value, and after " enrich- 
ment" (i.e. saturation with low-boiling, unsaturated hydrocarbons), 
is much used in this country as an illuminant (p. 312). 

It appears that carbon when burned with oxygen first forms car- 
bon dioxide with evolution of 97.0 Cal., but in the presence of excess 
carbon, the dioxide is partially reduced to monoxide, thus : 

C + CO 2 = 2 CO - 39 Cal. 

The lower the temperature the less complete is the reduction. By 
the mass action law, at any given temperature, the partial pressure 
of the carbon dioxide, divided by the square of that of the carbon 
monoxide, is at equilibrium a constant, which is, however, a function 

of the temperature; i.e. P 00 * = Ki = fi(f). In the table below 

(Pco)- 2 

the value of this constant for various temperatures is given, and it 
appears that a large reduction of dioxide with carbon is only possible 



at high temperatures. The proportion of carbon dioxide in the gases 
at a given temperature is generally greater than the figures indicate, 
since the rate of reaction between carbon dioxide and carbon is rela- 
tively low ; but it decreases rapidly with the temperature. If the 
gases are in contact with carbon for a sufficient time, at the given 
temperature, the ratio of dioxide to monoxide corresponds to the 
equation, independently of the presence of other gases. 


(Pressures in Atmospheres) 




























The mass action law for the water-gas reaction requires that 
H2 = KZ = fz(t) ; this second constant is also shown in the 

table above. From this it appears that at high temperatures only 
does the reduction of the steam approximate completion. Since the 
carbon monoxide formed by this reaction is necessarily accompanied by 
carbon dioxide corresponding to its equilibrium with carbon, it is seen 
that at high temperatures only will the gas coming from the producer 
be satisfactory, while at lower temperatures much undecomposed steam 
and carbon dioxide will be present. This is shown in the table below. 

(Equilibrium of Steam and Carbon) 



%H 2 

%H 2 

%C0 2 











* Haber, Thermodynamics of Technical Gas Reactions. 


The water vapor will largely separate on cooling the gas, but the 
carbon dioxide remains as an undesirable diluent. Furthermore, a 
commercial producer, especially when forced in capacity, owing to 
incomplete reactions, gives lower yields of monoxide and hydrogen 
than is shown ; it should not be run at less than 1000 C. 

In operation, the fuel is brought to white heat by an air blast, 
which is then cut off and steam injected ; the formation of water 
gas now proceeds until the temperature falls below 1000 C., owing 
to heat absorption ; then the steam is cut off, and the air blast turned 
on until the fuel is again incandescent, and the cycle of operations 
repeats. During the air blow, the gas formed is sent to the chimney 
but during the steam blow, the water gas is cooled and sent to the 
holder. As the fuel bed is deep and above 1000 C., the air blow 
forms a large amount of carbon monoxide, which may be collected for 
use ; but it is often run to waste, in which case the air supply should 
be large to secure a maximum formation of carbon dioxide, which is 
not in contact with the fuel a sufficient time for reduction to carbon 
monoxide. Even under the best conditions, a serious loss of heat re- 
sults from the incomplete combustion of the fuel.* 

The waste of energy of the coal in this intermittent process can 
be largely avoided by combining the two reactions ; the air and steam 
being regulated so the heat of combustion of the carbon with the air 
just compensates the heat of formation of the water gas ; the mixture 
of gases so obtained is called producer gas. The product contains 
much free nitrogen, but the simplicity of the process causes this arti- 
ficial gas to find use on a large scale. 

Since the fuel must be kept at about 1000 C. to secure complete 
reaction between the carbon and steam, and ensure low carbon 
dioxide content of the gas, the product leaving the generator carries 
very high sensible heat. In theory it is possible to recover this by 
using it to heat the air and steam supply, thus saving energy which 
must otherwise be furnished by burning more coal, but owing to the 
complications of the system, this heat recovery is not usually prac- 

The use of steam in a producer to lower the temperature results 
in smaller loss of sensible heat in the gas, and theoretically should 
increase the efficiency of the gas producer. But in fact the gain is 

* Anthracite produces a higher content of hydrogen and methane than does 
coke, but the latter yields some methane, due to the action of hydrogen on the 
carbon at the high temperature. The gas also contains some nitrogen from the air 
blast and often a little oxygen due to leaks. 


offset by the large heat loss due to undecomposed steam, which 
escapes reduction by the carbon, owing to the slow rate of reaction 
between steam and carbon; much heat is wasted in bringing this 
steam up to the temperature of the producer. Since lowering the 
temperature avoids clinkering of the ash and decreases the wear and 
tear on the apparatus, steam is, however, always used in excess. 

In the Siemens gas producer * (Fig. 21), the coal is introduced 
at (E), falls upon the step grate (B, B), and is brought to incan- 
descence by air entering through the openings while steam is injected 

FIG. .21. 

Fio. 22. 

from the pipe (C), and the gas formed escapes through (A, A). The 
ashes fall through the grate (G) into the pit, which is kept closed 
except when cleaning. A more modern producer (Taylor's) is shown 
in Fig. 22. The coal rests on a bed of ashes (A, A), and air is forced 
through the blast pipe (F), raising the fuel to incandescence. The 
gas formed passes out by the pipe (E). The grate (G) is made to 
revolve by the crank at (B), and the ashes fall over the edge of the 
grate at (H). The bed of ashes* is kept about 3 feet deep on the 
revolving bottom. Steam from the pipe (D) is introduced with 
the air through the blast pipe, which is provided with a hood to 
disseminate them through the fuel. In all plants burning producer 
gas the regenerative or recuperative heating system is used. 

* Jour. Soc. Chem. Industry, 1885, 441. 


Mond gas is producer gas made from coal slack and with a very 
large excess of steam in the blast. Much undecomposed steam passes 
out with the gas and the temperature in the producer is kept low 
(exit gas about 500) so that the major reaction taking place is, 

C + 2 H 2 = CO 2 + 2 H 2 . 

This avoids destruction of the ammonia content of the gas, which is 
higher than in water gas, and by scrubbing with water and sulphuric 
acid, ammonium sulphate is obtained as a by-product. The gas has, 
however, rather low calorific value (140 B. T. U. per cu. ft.) and be- 
cause of its high hydrogen content is not well suited for open-hearth 
steel furnaces. 

Blast furnace gas. The waste gases from iron blast furnaces 
contain about 30 per cent of carbon monoxide and 58 to 60 per cent 
of nitrogen. About one-half of the gas is used for heating the air 
blast and the rest under boilers, or in large gas engines to generate 
power for driving the blowers and other purposes. The gas has to be 
carefully purified from dust before delivery to the gas engine. 

The efficiency of the internal-combustion engine is much greater 
than that of the steam engine ; thus the use of producer gas in gas 
engines is the most efficient method of converting the energy of 
coal into power. But this high energy efficiency is largely offset 
by the greater cost of installation and maintenance, and by lack of 
" overload " capacity of the engine. Thus with the exception of 
conditions of unusual constancy of load, or of very high cost of fuel, 
the steam plant is the cheapest method of developing power. 

To secure high temperatures by gas combustion, preheating of 
both gas and air before they enter the combustion chamber is neces- 
sary. This avoids the cooling effect they otherwise exert, and is 
most economically done by recovering the waste sensible heat in 
the reaction products, by means of the Siemens regenerative furnace, 
which is shown in its simplest form in Fig. 23. The material to be 
heated is placed on the furnace hearth (A). Four passages (B, C, 
D, and E), filled with loosely piled fire-brick are called the " checker 
work." On their way to the chimney, the hot gases from the furnace 
pass through and heat two checker works, e.g. (B) and (C). When 
they are sufficiently heated, the flow of furnace gases is turned into 
(D) and (E), through which they pass to the chimney. Then fuel gas 
is conducted through the hot passage (B), to the furnace (A), where 
it mixes with air which has been heated by passing through (C). 
The temperature of (A) is thus much higher than if the air and gas 



arrived at (A) cold. While (B) and (C) are being thus cooled, (D) 
and (E) are heated by the waste gases, and after a time the dampers 
are turned, the gas made to pass through (E), and the air through 
(D), while the combustion products pass through (B) and (C) to the 
chimney. Hence the process is an alternating one, the checker works 

FIG. 23. 

on one side being heated, while those on the other are giving up their 
heat to the gas and air respectively. The interstices between the 
bricks of the checker work often become clogged with ashes and soot. 
Recuperative heating involves the counter current flow of the hot 
combustion gases through or around flues, on the other side of which 
the cold gas and air pass to the furnace. This method has the disad- 
vantage of low heat conductivity of the flue walls, but avoids all in- 
termixture of the in-coming and out-going gases. The flow of the 
gases through the recuperator is continuous and there is no periodic 
changing of valves as in the other system. 

An average composition of various fuel gases is as follows : 

H 2 


C 2 H 6 


C0 2 

N 2 


Natural gas (Ohio) * . . 
Coal gasf 
Water gasf 









Producer gas (coal) f . . 







When burned with 20 per cent excess of air, and assuming that 
the escaping gases have a temperature of 500 F., 1000 cubic feet 

* Gas and Fuel Analysis for Engineers, A. H. Gill. 

t Industrial Chemistry, Rogers and Aubert, 2d ed., p. 404. 


of gas will evaporate the following number of pounds of water, at 
from 60 F. to 212 F. : - 

Natural gas 893 pounds * 

Coal gas 591 pounds 

Water gas " . . . 262 pounds 

Producer gas 115 pounds 


Liquid Fuel. B. H. Thwaite, London, 1887. (Spon.) 

Chemical Technology. Groves and Thorp. Vol. 1, Fuel, by Mills and 

Rowan, Phila., 1889. (Blakiston.) 
Feuerungsanlagen. F. Fischer, Karlsruhe, 1889. 
Liquid Fuel. E. A. B. Hodgetts, London, 1890. (Spon.) 
Die Feuerung mit flussigem Brennmaterialien. I. Lew, 1890. 
Fuels. H. J. Phillips, London, 1891. 
Fuels. C. W. Williams and D. K. Clark, London, 1891. 
Die Chemie der Steinkohle. F. Muck, Leipzig, 1891. (W. Engelmann.) 
Die Gasfeuerungen fur metallurgische Zwecke. A. Ledebur, Leipzig, 


Taschenbuch fur Feuerung stechniker. F. Fischer, Stuttgart, 1893. (Enke.) 
Contribution a 1'etude des combustibles. P. Mahler, Paris, 1893. 
Die chemische Technologic der Brennstoffe. F. Fischer, Braunschweig, 

A Treatise on the Manufacture of Coke and the Saving of By-Products. 

John Fulton, Scranton, Pa., 1895. 

Mineral Industry, 1895, 215, W. H. Blauvelt. (By-Product Coke Ovens.) 
Grundlagen der Koks-Chemie. Oscar Simmersbach, Berlin, 1895. 

(J. Springer.) 
The Calorific Power of Fuels. Herman Poole, New York, 1898. (Wiley 


Gas and Fuel Analysis for Engineers. A. H. Gill, New York, 3d ed., 1902. 
Modern Power Gas Producer Practice. H. Allen, 1908. 
Kraftgas. Dr. F. Fischer, Leipzig, 1911. (Otto Spamer.) 
Oil Fuel. S. H. North and Ed. Butler, 2d ed., London, 1911. 

* Orton, Geology of Ohio, Vol. VI, p. 544. 


Water for industrial use is chiefly obtained from : 

1. Surface waters, consisting of, 
(a) Flowing waters (streams). 
(6) Still waters (ponds, lakes). 

2. Ground waters, furnished by, 

(a) Springs. 

(b) Shallow wells (usually penetrating but one geological 


(c) Deep wells (passing through several geological strata). 
Sea water, aside from some use for condensers and for supply to 

marine boilers, finds but little direct use. Rain water collected from 
clean roofs or other surfaces furnishes excellent soft water in limited 
amount, but little is so obtained industrially. 

The impurities contained in water may be derived from the ground 
with which it has been in contact, or by contamination with sewage 
or factory wastes. In general the impurities in water constitute four 
classes : 

I. Dissolved gases, such as oxygen, carbon dioxide, hydrogen 
sulphide, etc. 

II. Soluble crystalloids, consisting of definite chemical com- 

pounds which cannot be removed by sedimentation or 

III. Soluble colloids, consisting of material of very high molecu- 

lar weight which will not settle or separate, and can be 
filtered only through semi-permeable membranes. 

IV. Suspended matter which will settle or can be filtered. 

I. Dissolved gases may comprise two groups : (a), those whose 
solubility diminishes sufficiently with increase of temperature in the 
solution, and which may thus be removed by merely heating the water 
in open heaters, or in closed vessels with the aid of vacuum; 
(b) those requiring chemical treatment. Acid gases may be neu- 
tralized (with calcium hydroxide or sodium carbonate), resulting in 
precipitation, or not, according to individual circumstances. Oxygen 
may be removed by passing the water over metallic iron, which 
is readily oxidized to the ferric state and precipitated, owing to 



II. The soluble crystalloids include most of the impurities occur- 
ring in natural waters, and their removal generally involves precipi- 
tation by chemical treatment. The presence of these substances 
in greater or less amounts imparts to the water those properties which 
render it hard, soft, saline, or alkaline. Some of these substances may 
be partially or wholly removed by merely heating the water. Here 
two groups are distinguished : (a) those which evolve a gas at higher 
temperatures and form insoluble bodies ; (6) those whose solubility 
decreases as the temperature increases in the solution ; thus hydrated 
salts in solution lose water of hydration as the temperature rises and 
pass into less hydrated and less soluble forms. 

The more common soluble crystalloids are the bicarbonates, sul- 
phates, and chlorides of calcium and magnesium ; sodium chloride, 
sulphate, and carbonate ; iron salts ; and silica. These are the sub- 
stances which cause the most difficulty in technical work, and espe- 
cially when the water is used in steam boilers. 

Hard water contains salts of calcium, magnesium, or iron, and is 
defined as one which precipitates soap from solution. Thus hardness 
is determined by titration with a standard soap solution. Tem- 
porary hardness is due to the presence of bicarbonates of iron and the 
alkaline earth metals ; the neutral carbonates are insoluble but dis- 
solve in water containing free carbon dioxide, forming CaH^ (003)2, 
MgH 2 (CO 3 ) 2 , etc. Permanent hardness is due to the presence of 
soluble neutral sulphates and chlorides of calcium and magnesium. 

Soft water usually contains very little mineral matter. Rain 
water as it falls is very soft, and if collected from clean surfaces is 
suitable for most purposes. Natural waters collected from ground 
containing little calcium or magnesium in soluble form is fairly soft 
as a rule; but if the water has percolated through soil containing 
peat or decaying vegetable matter, it is often discolored by dissolved 
organic matter, and may contain organic acids which cause corrosion 
of iron or other metals. 

Saline and alkaline waters contain the sulphates, carbonates, or 
halogen salts of the alkali metals, in rather large amounts. Sea water 
and many ground waters (springs, wells) are characteristically saline 
(mineral waters). Alkaline waters are high in carbonates and sul- 
phates ; as, e.g. the " alkali " waters of the western states. 

III. Colloidal substances have very large molecular weight, and 
are characterized by the tendency to adsorb or condense on the 
boundary surfaces of precipitated matter, if these surfaces are rela- 
tively large. Thus to remove colloidal matter from solution, a 


flocculent or finely divided amorphous precipitate may be produced 
in the water, which adsorbs the colloidal matter and attaches it to the 
precipitate. Crystalline precipitates have relatively small surfaces, 
and do not serve well for removing colloidal matter. 

Colloidal substances in water form two groups : (a) Suspension 
colloids, consisting of small particles of suspended solid matter, which 
may be easily coagulated and precipitated by adding some electro- 
lite. (6) Emulsion colloids consisting of minute particles of in- 
soluble matter probably liquid, suspended in the water, e.g. emulsified 
oils, gelatine, gums, and certain hydrated compounds mostly of complex 
structure, such as some varieties of clays, tannic acids, humus bodies, 
etc. Emulsion colloids are usually difficult to coagulate, and only 
when much precipitating agent is added, can it be done. Thus if 
emulsion colloids predominate in a water, purification may be imprac- 
ticable, owing to the large quantity of precipitant needed ; but these 
colloids can often be adsorbed on the surfaces of suitable materials, 
such as alumina, iron hydroxide, charcoal or bone-char. 

Some substances present in natural waters may impart to it 
corrosive properties, making its use objectionable for industrial 
purposes. Dissolved oxygen and carbon dioxide, hydrogen sulphide, 
free acids, either mineral or organic, and easily hydrolized salts, as 
MgCU and FeSO4, are very liable to cause corrosion or " pitting " 
in a boiler. Some of these may be derived from factory wastes, or 
from swamps and peat bogs, or from mine sumps and drainage from 
culm or refuse dumps. 

The purification of water for industrial use consists in the partial 
or complete removal of the objectionable substances suspended or 
dissolved in it. This is often difficult, owing to the nature of the 
impurities ; the size of the plant required for large works is also an 
item of concern. The quality of the water available should be con- 
sidered in locating the works. 

Water containing suspended matter only may be purified by sedi- 
mentation, followed by sand filtration; but this is often combined 
with chemical treatment, by which a precipitate is formed in the water. 
This precipitate acts mechanically to entangle the suspended matter ; 
and it also acts as an adsorption agent on emulsion colloids, such as 
dissolved organic coloring matter, grease or oils, glutinous substances, 
and many kinds of factory wastes. Aeration by spraying into the 
air, or by trickling in thin films over large surfaces, accelerates the 
escape of carbon dioxide, or hydrogen sulphide, while absorption of 
oxygen aids precipitation of iron from solution. Bacteria and other 


organisms are frequently destroyed in sewage effluents, and in mu- 
nicipal supplies, by treatment with hypochlorites, ozone, or copper 
sulphate. Generally the raw water is mixed with some soluble salt, 
such as aluminum or ferrous sulphate, which is precipitated as alumi- 
num or iron hydroxide by the action of the alkaline substances in the 
water, or added to it later. This gelatinous precipitate encloses 
suspended matter, and combines with soluble organic coloring matter 
by adsorption; by filtration on sand filters (p. 19) it is removed, 
carrying with it the impurities. But this increases the soluble im- 
purity by the alkaline sulphates left in the water. Sulphuric acid 
and iron sulphate are removed from Allegheny River water for the 
Pittsburgh city supply, by adding calcium chloride.* The calcium 
sulphate precipitate aids in removing the suspended silt and coloring 

Water for boiler supply is generally treated to reduce the hard- 
ness or to neutralize its corrosive properties ; the operation is called 
" softening.'* Temporary hardness is removed by some of the fol- 
lowing methods : 

1. Boiling the water, usually in "feed-water heaters," to decom- 
pose the bicarbonates : 

CaH 2 (CO 3 ) 2 = CaCO 3 + H 2 O + CO 2 . 

Feed-water heaters are heated by exhaust steam, or waste flue-gases, 
and may be " open," when working under atmospheric pressure, or 
" closed " if under internal pressure (as economizers), or under 
vacuum. Open heaters permit the ready escape of dissolved gases 
and decomposition of bicarbonates, with precipitation of iron, cal- 
cium, and magnesium, but for complete separation of the alkaline 
earths, a small amount of sodium carbonate must be added to the 
water in the heater, to decompose any permanent hardness. Closed 
heaters working under pressure (economizers) afford less complete 
separation of hardness, since the gases cannot readily escape. In 
vacuum heaters the gases are removed, and the water is purified. 

Heaters employing waste flue gases are known as " economizers," 
the water being heated in tubes set in the furnace flue. The condi- 
tions are essentially those prevailing in the boiler and the scale de- 
posits inside of the tubes, which need to be frequently cleaned. 

2. Treatment with calcium hydroxide (" milk of lime ") : 

CaH 2 (C0 3 ) 2 + Ca(OH) 2 = 2 CaCO 3 + 2 H 2 O. 

* Hoffmann, J. Ind. Eng. Chem., VI (1914) p. 52. 


The clear calcium hydroxide solution obtained by letting the undis- 
solved lime settle is preferable for this, but frequently unsettled 
" milk of lime " is used. The required amount of quicklime is 
slaked in a little water, and the " milk " thoroughly mixed with the 
water to be purified. This is Clarke's process. The sludge of cal- 
cium carbonate is removed by settling in suitable tanks, or by a 

3. Treatment with sodium carbonate (soda-ash) : 

CaH 2 (CO 3 ) 2 + Na 2 CO 3 = CaCO 3 + 2 NaHCO 3 . 

The permanent hardness is less easily remedied, for in these cases 
treatment of the water leaves some substance more or less deleterious 
in solution : 

1. CaSO 4 + Na 2 CO 3 = CaCO 3 + Na 2 SO 4 . 

2. CaS0 4 + Ba(OH) 2 = BaSO 4 + Ca(OH) 2 . 

3. CaCl 2 + Na 2 C 2 O 4 = 2 NaCl + CaC 2 O 4 . 

Much care is necessary to avoid an excess of the chemical added.* 

When natural water containing soluble impurities is used in a 
boiler, a more or less coherent deposit, called boiler scale, forms on 
the plates and tubes. This is chiefly composed of carbonate and sul- 
phate of calcium ; but in some cases magnesium hydroxide and sul- 
phate, iron hydroxide or oxide, silica and organic matter, are present. 
The decomposition of bicarbonates of calcium and magnesium by 
heat, as above indicated, also takes place in the boiler. Calcium car- 
bonate alone forms a porous, non-adherent scale or sludge, which is 
largely removed by " blowing off " or washing out the boiler. Cal- 
cium sulphate is rendered less soluble by the heat and pressure within 
the boiler, and is deposited as a hard, compact scale, adhering firmly 
to the plates and tubes. Magnesium sulphate, if present, is deposited 
as monohydrated salt (MgSO 4 ' H 2 O), and is strongly adherent. 

Scale formation is very detrimental ; being a poor conductor of 
heat, the evaporative capacity of the boiler is reduced and much 

* Besides the methods given above, many other substances have been proposed, 
and are used to some extent, usually within the boiler itself, for water purification. 
Among these are sodium hydroxide, phosphate, aluminate, fluoride, oxalate, silicate, 
and bichromate ; also barium hydroxide and aluminate. But in general these are 
too expensive for large works. 

1. MgH 2 (CO 3 ) 2 + 2 NaOH = MgCOs + Na 2 CO 3 + 2 H 2 O. 

2. CaH 2 (CO 3 ) 2 + Na 2 Cr 2 O 7 = CaCrO 4 + Na 2 CrO 4 + 2 CO 2 + H 2 O. 

3. MgSO* + Na 2 Cr 2 7 = MgCKX + Na 2 SC>4 + CrO 3 . 
J. Am. Chem. Soc. 1899, 655. Eng. Min. Jour. LX, 220. 


more fuel is consumed. The scale separates the water from the boiler 
plates and tubes, which thus are overheated and rapidly burn out. 
The tubes also become clogged and their efficiency is much impaired. 
Magnesium chloride is especially troublesome, for it not only 
forms scale, but causes rapid corrosion of the iron, possibly thus : 

MgCl 2 + Fe + 2 H 2 O = Mg(OH) 2 + FeCl 2 + H 2 . 

Dissolved oxygen and carbon dioxide, and hydrogen sulphide, are 
strongly corroding, the latter probably because of oxidation to sul- 
phuric acid. Manganese sulphide in the boiler plates assists depo- 
larization of the hydrogen ; it may also be oxidized and hydrolized 
to form sulphuric acid, and thus accelerate corrosion. 

As a rule the water should be treated before it goes into the boiler, 
but if the scale-forming impurity does not exceed 170 parts per mil- 
lion, the purification may be done in the boiler itself, followed by a 
daily "blowing off." A good circulation of water in the boiler tends 
to keep the precipitated matter loose so it may be easily blown out. 

In some cases scaling may be prevented by introducing colloidal 
substances into the boiler ; possibly the particles of incrusting matter 
become coated with a thin film of the colloid by adsorption. Coales- 
cence of the particles of the crystalloids is thus prevented and they 
form a loose sludge, or are kept in suspension in the water, until re- 
moved by " blowing off," or w r ashing out the boiler. The use of 
kerosene, of tannins, and of other organic substances in the boiler is 
based upon this action. 

Many proprietary " anti-scale " preparations are sold, some of 
which are of no particular value. These are generally intended for 
use inside of the boiler, and may act by direct precipitation of mineral 
matter, or by preventing adhesion to the boiler plates. They often 
contain soda-ash, caustic soda, sodium silicate or phosphate, barium 
hydroxide, tannins or vegetable extracts, or petroleum products. 

Saline or alkaline waters, or those whose content of alkali sulphates 
and chlorides has been artificially raised by purification methods 
as above, may give trouble in steam boilers by causing "priming," 
i.e. the passage of water particles, mixed with the steam, from the 
boiler. Priming is associated with foaming, resulting from much 
dissolved matter, or due to finely divided suspended particles. When 
water contains a great number of suspended fine solid particles, each 
serves to release steam bubbles in its immediate vicinity, and this 
increases the space occupied by the water in the boiler, i.e. foaming is 



caused.* The presence of large amounts of alkaline chlorides and 
sulphates, along with much calcium or magnesium sulphate, may ren- 
der a natural water useless for steam raising, since it will foam after 
chemical treatment. No satisfactory method of purifying such waters 
has yet been proposed. Frequent blowing off of the boiler is the only 
preventive of foaming. In locomotives foaming may occur when 
the dissolved salts amount to 1700 parts per million ; but stationary 
boilers will permit four times as much. 

Grading of natural water for steam raising is often difficult, owing 
to the local conditions in the region where the water is to be used. A 
water containing 250 parts per million would be considered poor in 
some parts of New England, but would rank as good in Dakota or 
Iowa. A classification proposed f for locomotive supply is shown 
below : 



More than 

Not more than 

More than 

Not more than 

Good . . . * . 







Fair . . . . . . 
Poor . . ' . 


Very bad . . , ,, 

The character of the water available is very important in some 
manufacturing processes. Hard water is objectionable for laundries, 
bleacheries, and soap works, since the insoluble lime soaps are pre- 
cipitated, causing loss of considerable soap, and injury to the goods, 
owing to the insoluble soaps adhering to them. 


The permutite process t for water softening is used somewhat in dye 
works and bleacheries. Permutite is an artificial zeolite (hydrated silicate 
of sodium and aluminum) made by fusing together feldspar, kaolin, and 
alkali carbonates, and lixiviating the pulverized mass with hot water. 
The composition is given as 2 SiO 2 -Al 2 O3-Na2O-6 H 2 0, and the substance 
is practically insoluble in water. Upon contact with water containing 
bicarbonates or sulphates of calcium or magnesium, or iron or manganese 

* Railroad Gazette, Oct. 12, 1900. C. H. Koyl. 

fProc. Am. Ry. Eng. and Maintenance of Way Assoc. V (1904), 595; IX 
(1908), 134. 

I Textile World Record, Nov. 1912. 


salts, an interchange takes place between the sodium of the permutite 
and the metal of the dissolved salts ; thus : 

Na 2 O A1 2 3 2 SiO 2 6 H 2 O + CaH 2 (CO 3 ) 2 = 

CaO A1 2 O 3 2 Si0 2 6 H 2 + 2 NaHCO 3 . 
NasO A1 2 3 2 Si0 2 - 6 H 2 O + MgSO 4 = MgO A1 2 3 - 2 SiO 2 . 6 H 2 O + Na2SO 4 . 

Thus by simple nitration through a layer of granular sodium permutite, 
a hard water can be softened. When the permutite becomes inactive 
through deposition of alkaline earth metal and removal of the sodium, the 
material can be regenerated in situ, by washing for 8 to 10 hours with a 
10 per cent sodium chloride solution ; thus : 

CaO A1 2 O 3 2 Si0 2 6 H 2 O + 2 NaCl = Na^O A1 2 O 3 2 SiO 2 - 6 H 2 O + CaCl 2 . 

Hard waters may cause unevenness of color deposition in dyeing ; 
they also lower the quantity of extract matter taken up from malt 
in brewing and distilleries. Iron in any form is very bad for dye- 
ing, tanning, paper making, bleacheries, laundries, and for brew- 
ing; it causes dark color in the material. Moderate hardness due 
to sulphates is of advantage in paper making; for tanning sole 
leather, owing to its swelling effect on the hides, and for brewing 
pale beers, since it decreases the extraction of coloring matters and 
protein substances from the malt. Chlorides are injurious in nearly 
all cases where the water comes in direct contact with the material, 
as in foods, in the brewing and distilling industries, in sugar mak- 
ing, and in tanning. 

Suspended matter, either of mineral or organic nature, and dis- 
solved organic coloring substances, may cause discoloration, or spots 
in goods with which they come in contact : thus paper-mills, dye works, 
bleacheries, and starch factories require a perfectly clear and color- 
less water. The presence of organic dirt may cause decomposition 
or putrefactive changes in food products, in the fermentation indus- 
tries, in starch and sugar making. 


Die Chemische Technologic des Wassers. F. Fischer, Braunschweig, 


Die Verhutung un'd Beseitigung des Kesselsteins. W. Storck, Halle, 1881. 
A Treatise on Steam Boiler Incrustations. C. T. Davis, Washington, 


Water Supply. William R. Nichols, 1886. 
Report on Boiler Waters of the C. B. & Q. R.R. W. L. Brown, Chicago, 


Die Verunreinigung der Gewasser. K. W. Jurisch, Berlin, 1890. 
Das Wasser. F. Fischer, Berlin, 1891. (J. Springer.) 
Das Reinigen von Speisewasser fur Dampfkessel. A. Rossel, Winterthur. 



L'Eau dans 1'Industrie. P. Guichard, Paris, 1894. (Bailliere.) 

Report of the Filtration Commission of the City of Pittsburg, 1899. 

American Machinist, 22 (1899). A. A. Gary. (Use of Boiler Com- 

L'Eau dans 1'Industrie. De la Coux, Paris, 1900. 

Boiler-Waters. Wm. W. Christie, New York, 1906. (Van Nostrand.) 

Clean Water and How to Get It. Allen Hazen, New York, 1907. 

Stream Pollution in Potomac River Basin. H. N. Parker. U. S. Geol. 
Survey, Water Supply Paper, No. 192 (1907). 

Disinfection of Sewage and Sewage Filter Effluents. E. B. Phelps. 
U. S. Geol. Survey, Water Supply Paper, No. 229 (1909). 

Proc. Western Railway Club, 1903, 241. 

Eng. Min. Jour. 1895, 220. F. Wyatt. 1899, 443. J. H. Parsons. 

J. Soc. Chem. Ind. 1884, 51. J. H. Porter. 1886, 267. Macnab and 
Beckett. 416. A. Steiger. 1887, 178. V. C. Driffield. 1888, 795. 
A.H.Allen. 1891,511. Archbutt and Deeley. 

Eng. News. 60 (1908), 355. H. Stabler. 


Most of the sulphur used in the industries is derived from the 
native mineral, which is found in many places, but usually in volcanic 
regions. It is always impure, being mixed with gypsum, aragonite, 
clay, or other matter, in the interstices of which the sulphur is de- 
posited. The formation of sulphur beds may have occurred by the 
reaction of gases, such as hydrogen sulphide and sulphur dioxide, 
with each other or with oxygen ; or by the decomposition of metallic 
sulphides through the agency of heat ; or by the reduction of sulphates, 
especially of calcium sulphate, which has probably caused the for- 
mation of some stratified deposits. 

The first is probably the most frequent mode of deposition, and 
may be observed at the present time in many volcanic districts where 
hydrogen sulphide and sulphur dioxide are escaping. The reactions 
are the following : 

SO 2 + 2H 2 S = 2 H 2 O + 3S; 

H 2 S + O = H 2 O + S ; 
H 2 S + 3 O = H 2 O + S0 2 . 

The largest part of the world's supply of sulphur comes from 
Sicily, but some is obtained in Japan, Italy, Greece, and in the 
United States, particularly in Louisiana, in Wyoming, in Utah, and 
near Humboldt, Nevada. The Louisiana deposit is now yielding 
a sufficient supply for our domestic consumption and shipment 
abroad has been introduced. 

In Sicily it is disseminated through the matrix, sometimes in 
considerable masses of nearly pure sulphur, but usually in fine seams 
or grains. The methods of obtaining it are very crude and wasteful. 
The mines are for the most part open pits, ranging from 200 to 500 
feet in depth, and the ore is carried to the surface in baskets or sacks 
by laborers, who ascend by inclined paths on the walls of the pit. 
In some of the better mines, however, hoisting machinery is now used, 
but only after overcoming the determined opposition of the laborers. 
The ore is generally refined in a very simple manner, the process being 
carried on in kilns called " calceroni" As usually constructed, these 
are shallow pits, about 30 feet in diameter, with walls about 10 feet 
high, made tight with mortar. They are generally built on a hill-side, 
and the sloping bottom is beaten smooth. The ore is arranged in 
the calcerone so as to leave a few vertical draught holes from top to 



bottom of the heap, which is fired by dropping burning brush or straw 
into these openings. The sulphur, forming from 25 to 40 per cent of 
the ore, burns freely, and when the heap is well on fire, the draught 
holes are closed, the calcerone covered with spent ore, and the whole 
left for several days. The heat given out by the burning of part 
of the sulphur is sufficient to melt the remainder from the gangue, 
and it collects in a pool near a tap-hole, made in the wall at the lowest 
point. At intervals of a few hours, the melted sulphur is drawn off 
into moulds. If the temperature rises above 180 C., there is a 
large formation of plastic sulphur, which will not flow from the tap- 
hole. To burn out a calcerone requires 35 to 80 days, depending on 
its size, the amount of gypsum in the ore, and the weather. From a 
quarter to a third of the sulphur is lost as sulphur dioxide, and as 
this damages vegetation, calcerone burning is prohibited during the 
spring and summer months. 

Of recent years the Gill kiln, paterned after the Hoffmann furnace 
(p. 185), has been introduced. This uses part of the sulphur as fuel 
and consists of four to six chambers, the air for combustion entering 
that chamber where the melting has just been finished. The air 
thus warmed enters the second chamber, where combustion is at its 
highest point and the sulphur is melted out of the ore ; the hot gases 
pass to the next chamber, filled with fresh ore, which is thus heated 
to the fusion point of the sulphur before combustion begins. The 
waste gases finally pass to the chimney. The yield of sulphur is 
considerably better than by the calcerone method. 

Processes for extraction of the sulphur by means of carbon disul- 
phide or other solvents have proved too expensive for industrial 
use. Extraction with superheated steam* yields an excellent quality 
of sulphur without formation of sulphur dioxide, and is used in this 
country and Japan to some extent. 

In Wyoming, sulphur occurs as irregular deposits or pockets in 
limestone. The ore is broken to small size, loaded into steel cars 
having perforated sides, and run into a retort where steam at 60 Ibs. 
pressure is admitted ; the sulphur melts, flows to the bottom of the 
retort, leaving the gangue rock in the car. 

By boiling the ore in a concentrated solution of calcium chloride f 
at 125 C., the sulphur can be melted from the gangue, with no forma- 
tion of sulphur dioxide, and no nuisance is caused. This has been 
tried in Sicily, but is not in general use. 

* J. Soc. Chem. Ind. 1887, 439, 442 ; 1889, 696. 

t Vincent, Bull. Soc. Chim. 40, 528. Am. Chem. Jour. VI, 63. 


In Louisiana, sulphur is obtained by the method devised by Her- 
mann Frasch,* which has been very successful. Driven wells are 
sunk into the deposit, which lies at a depth of about 450 feet, and is 
about 100 feet thick. In each well are four concentric lines of pipe, 
ranging in diameter from 10 inches to 1 inch. Superheated water 
(165 to 170 C.) is forced down between the 10-inch and 6-inch pipes, 
and passing into the crevices of the sulphur-bearing rock, melts the 
sulphur, which runs into the sump at the foot of the well. Through 
the 1-inch pipe, compressed hot air is forced to the bottom of the well, 
where it mixes with the melted sulphur, forming an aerated mass, 
which the water and air pressure cause to rise through the 4-inch pipe, 
to the surface ; the mixture of melted sulphur, hot water, and air is 
discharged into large open vats made of boards. The solidified sul- 
phur goes direct to market without further refining, and is of better 
quality than that from Sicily, which it has practically displaced from 
the American market. 

A small part of the sulphur of commerce is recovered sulphur, 
chiefly obtained from the calcium sulphide waste of the Leblanc 
soda process (p. 103 et seq.), and from the residues from the purifica- 
tion of illuminating gas by iron oxide (p. 321). The residues con- 
taining 50 to 60 per cent of free sulphur are heated in a retort and the 
sulphur distilled off. It seems improbable that recovered sulphur 
will ever become much of a factor in the market. 

The sulphur obtained by the above described processes is gen- 
erally pure enough for manufacturing and agricultural uses. But in 
a few industries, a refined sulphur is needed. This is produced by 
distillation from an iron retort, the vapors being condensed in brick 
chambers. If the temperature of the chamber is not above 110 C., 
the vapors condense at once to a fine powder, called " flowers of 
sulphur " ; but if the temperature in the condensing chamber rises 
much above 110 C., the vapors condense to a liquid, which is drawn 
into moulds to form the " roll brimstone " of commerce. 

The chief uses for crude sulphur are: for combating Oidium 
tuckeri, a fungus causing the vine disease (this disposes of a large part 
of the yearly production) ; for making sulphuric acid ; for sulphurous 
acid and bisulphite solutions ; for carbon di sulphide ; and for making 
ultramarine. Refined sulphur goes mainly for gunpowder, matches, 
and for vulcanizing rubber. 

* U. S. Pat. Nos. 461429, 461430, 461431. Mining World, 1907, 1049. 
Eng. Min. J. 1907 (84), 1107. Mineral Resources of the United States, 1907, 
(Pt. II), 674, 



Sulphur melts at 115-120 C., and has a specific gravity of 1.98- 
2.04 ; it is a poor conductor of heat and electricity, dissolves easily 
in carbon disulphide, and less readily in chloroform, benzol, turpentine, 
and other oils. 

Sicily, owing to its favorable situation as a shipping point, the 
abundance of cheap labor, and its rich deposits, was long the dominat- 
ing factor in the sulphur market. But recent competition with the 
American and Japanese production of sulphur has caused the closing 
of many of the mines, and those which have continued working have 
met with large losses. Only by greatly improving their methods of 
mining and refining the ore can the industry be restored to a satis- 
factory condition. At the present time the American sulphur industry 
in Louisiana is in a flourishing condition. 


Sulphur dioxide (SO 2 ) is the most important sulphur compound 
and is made on a large scale by roasting iron pyrites (p. 66) for the 

sulphuric acid manufacture : for pro- 
ducing smaller quantities of sulphur 
dioxide direct combustion of brim- 
stone is customary. Brimstone burn- 
ers may be simple brick ovens, or 
long iron retorts, in which the sul- 
phur is ignited and a regulated 
supply of air admitted to ensure 
complete combustion. Too much 
heat in the retort may cause distil- 
lation of sulphur into the flues and 
other parts of the apparatus. Me- 
chanical sulphur burners afford more 
uniform combustion and high con- 
centration of the sulphur dioxide gas. 
A modern type is the Wise sulphur burner (Fig. 24), consisting of a 
cast-iron bowl (B), with an agitator (A) whose ploughs dip into the 
melted sulphur (C) in the bowl. Air enters by openings (D) in the 
wall just above the surface of the burning sulphur. In the annular 
trough (E) is placed brimstone, which is melted by the heat of the 
burner itself, and flows through the inlets (F) into the bowl, replacing 
that which is burned to sulphur dioxide. Above the sulphur pot is 
a combustion box (G), in which is a horizontal baffle plate with slots 

FIG. 24. 


at each end. Opposite each slot is a damper in the end of the com- 
bustion box, to admit air for completing the combustion of any sul- 
phur vapor. Thus sublimation of the sulphur is avoided and the 
sulphur dioxide concentration in the gas is high, 18 to 19 per cent by 
volume being claimed. 

The Tromblee-Paull burner is a rotary, horizontal iron cylinder, 8 
feet long and 3 feet in diameter, having conical ends ; it makes one revo- 
lution in two minutes, and consumes about 5500 Ibs. of sulphur per 
day. The sulphur, melted by the heat of the burner, flows into the 
cylinder, coating the interior of the shell in consequence of the rotation. 
Air entering through suitable dampers in one end of the cylinder burns 
the sulphur on the inner surface of the shell, and the gases pass into a 
combustion chamber to complete the burning of any volatilized sulphur. 

Much sulphur dioxide is produced in the roasting and smelting 
of copper and lead ores (p ; 594 et seq.) and recently attention has 
been given to the condensation of these fumes for the making of acid 
or other purposes, but chiefly with the object of abating the nuisance 
and damage they cause in the surrounding country. 

Pure sulphur dioxide is made by dissolving the crude gas from 
sulphur burners in water by use of counter-current washing towers, 
and recovering it from the solution by heating. The gas is dried, 
compressed to liquid, and put on the market in steel cylinders. 

Sulphur dioxide is used for making sulphuric acid ; for the acid 
sulphite liquor used in making wood pulp ; for preparing sodium bisul- 
phite as a bleaching agent for wool, hair, straw, and other tissues ; 
as a disinfectant and germicide ; and in the liquid state in ice machines. 

Substances such as wool and straw, when bleached by exposure 
to sulphur dioxide gas, slowly regain their original color on exposure 
to the light. The coloring matter is not destroyed, but probably 
unites with the sulphur dioxide to form a colorless compound, which 
slowly decomposes. 

Sodium bisulphite (NaHSO 3 ) is formed by saturating sodium car- 
bonate solution with sulphur dioxide : 

Na 2 CO 3 + H 2 O + 2 SO 2 = 2 NaHSO 3 + CO 2 . 

It forms a- strong-smelling solution occasionally used as an " anti- 
chlor " to remove excess of chlorine from the fibres of bleached cotton 
or linen goods. Its reaction is probably as follows : 

Ca(ClO) 2 + 2 NaHSO 3 = 2 NaCl + CaSO 4 + H 2 SO 4 ; or, 
= Na 2 SO 4 + CaSO 4 + 2 HC1 ; 

2 Cl + NaHSO 3 + H 2 O = NaCl + H 2 SO 4 + HC1 ; or, 
= NaHSO 4 + 2 HC1. 


It also finds some use in other industries, such as chrome tannage, 
brewing, glucose and starch making. The solution of bisulphite 
decomposes on evaporation, giving off part of the sulphur dioxide, 
and forming neutral sulphite of sodium. 

Calcium bisulphite [CaH^SOs^] is made by passing sulphur di- 
oxide into milk of lime. It is probably a solution of neutral sulphite 
in an excess of aqueous sulphurous acid. It is used in much the same 
way as the sodium salt. 

Hydrosulphurous acid (H 2 S 2 O 4 ) and sodium hydrosulphite are 
important bleaching and reducing agents. The acid results from the 
action of iron or zinc on aqueous sulphurous acid : 

2 H 2 SO 3 + Zn = ZnO + H 2 S 2 O 4 + H 2 O. 

The zinc oxide unites with another molecule of sulphurous acid, form- 
ing zinc sulphite. 

Sodium hydrosulphite (Na 2 S 2 O 4 )* is made by dissolving zinc in 
sodium bisulphite : 

4 NaHSO 3 + Zn = Na 2 Zn(SO 3 ) 2 + Na^O, + 2 H 2 O. 

The zinc-sodium sulphite is precipitated with milk of lime, and the 
hydrosulphite is left as a solution, which is very unstable, rapidly 
absorbing oxygen from the air. The hydrosulphite can be precipitated 
from the solution by adding salt, and cooling the liquid, when crys- 
tals of Na 2 S 2 O 4 2 H 2 O separate. By treatment with hot alcohol, 
the water of crystallization is removed, and the anhydrous powder is 

SO 2 Na 

fairly stable if kept dry; its formula is O .' It can also be 


made by treating metallic sodium with sulphur dioxide (gas or liquid) 
in the presence of petroleum ether. 

Hydrosulphite combines with formaldehyde, producing a stable 
mixture of formaldehyde compounds of sodium sulphoxylate, and 
sodium bisulphite : 

SO 2 Na SONa SO 2 Na 

O +2CH 2 O + H 2 O = +0 


(sulphoxylate) (bisulphite) 

*Bernthsen, Ber. 13, 2277; 14, 438; 33, 126. 


This mixture is much used under various trade names, hydrosulphite 
NF, rongalite C, hyraldite, decroline, blanchite, etc., as reducing, bleach- 
ing, and discharge agents in textile industries. By treatment with 
zinc dust, the bisulphite-formaldehyde is converted into the sulph- 
oxylate-formaldehyde (NaHSO 2 CH 2 O - 2 H 2 O), which is sold as 
hydrosulphite NF, cone. ; hyraldite C, extra, etc. 

Sodium thiosulphate (Na 2 S 2 O 3 5 H 2 O), sold under the trade name 
" hyposulphite of soda" is made by digesting sulphur with a solution 
of neutral sodium sulphite, or sodium hydroxide : 

Na 2 SO 3 + S = 

6 NaOH H- 12 S = Na^Og + 2 Na 2 S 5 + 3 H 2 O. 

It is also obtained from the waste sulphide liquors of the Leblanc 
soda process (p. 104). It is largely used in chrome tannage, in photog- 
raphy, in wet silver-extraction processes, as antichlor in paper bleach- 
ing (p. 562), in textile dyeing and printing, for bleaching straw, wool, 
ivory, etc., and in iodometry. 


Sulphuric acid is probably the most important of all chemicals, 
because of its extensive use in a very large number of manufacturing 
operations. Of the immense quantities made yearly, the greater part 
does not come upon the market; for, being expensive and difficult 
to ship, consumers of large amounts generally make their own acid. 

The commercial grades of acid have special names. A moder- 
ately strong acid (50-55 Be.), such as condenses in the lead cham- 
bers, is known as " chamber -acid." It contains from 62 to 70 per 
cent of H2SO4, and is strong enough for use in the manufacture of 
fertilizer, and for other purposes requiring a dilute acid. By con- 
centrating this chamber acid, an acid of 60 Be. is obtained, contain- 
ing about 78 per cent of H 2 SO 4 , which is sufficiently strong for most 
technical uses. Further evaporation in platinum or iron pans yields 
an acid of 66 Be., containing 93.5 per cent of H^SO^ and known as 
oil of vitriol, while the strongest acid that can be made by direct 
evaporation contains about 98.5 per cent of H 2 SO 4 , and is called 
monohydrate. Fuming or Nordhausen acid, which is still more con- 
centrated, is prepared by special means, and it is essentially a solu- 
tion, of sulphuric anhydride (SOs) in sulphuric acid ; this is the acid 
which was prepared by the alchemists in the Middle Ages. 

In about the year 1740, Ward, an Englishman, began to make 
sulphuric acid on a moderately large scale. He burned sulphur and 
nitre (KNOs) together, and condensed the vapors in glass vessels 
containing a little water. The dilute acid so formed was then con- 
centrated in glass alembics or retorts. In this way an acid was 
produced at a lower price than the fuming acid could be made, and 
the industry was soon established on a commercial scale. 

Sulphuric acid is now made by two important methods : the old 
chamber process yielding dilute chamber acid (p. 74) directly, and 
the newer contact processes yielding sulphuric anhydride (SOs) 
as first product, from which any desired strength of sulphuric acid 
may be made by dissolving in weak acid or water. For producing 
concentrated acid the contact method has proved generally more 
economical, and is slowly displacing the old chamber process with its 
concentrating plant. But for acid of 50-60 Be., the advantage is not 
so decidedly in favor of the newer method. It is probable that the 
lead chamber will not be entirely given up for many years to come. 



The reactions involved in Ward's process are those of the present 
chamber process for sulphuric acid. This consists in bringing to- 
gether, under suitable conditions, sulphur dioxide, oxygen, and water 
vapor, in the presence of certain oxides of nitrogen. The latter 
probably act catalytically, causing the oxygen to unite with the sul- 
phur dioxide and water to form acid. The apparent reaction is : 

SO 2 + H 2 O + O = H 2 SO 4 . 

But this does not represent the actual process, which is more 
complicated than it at first appears. Several theories have been 
advanced to explain the reactions occurring in the lead chambers, 
and the part taken by the nitrogen oxides, but the most generally 
accepted one, that of Lunge, regards nitrous anhydride (N 2 O 3 )* as 
the essential factor, f According to this view, the principal reactions 
involved are as follows : 

1) 2 S0 2 + N 2 O 3 + O 2 + H 2 O = 2 SO 2 (OH) (ONO) (Nitrosyl- 
sulphuric acid) ; 

2) 2 SO 2 (OH) (ONO) + H 2 O = 2 SO 2 (OH) 2 + N 2 O 3 ; or, 

3) 2 SO 2 (OH)(ONO) + SO 2 + O + 2 H 2 O = 3 SO 2 (OH) 2 + N 2 O 3 . 

First there is a union of sulphur dioxide, nitrous anhydride, 
oxygen, and water, to form nitrosylsulphuric acid, which probably 
separates as part of the mist or fog seen in the lead chambers. But 
in the presence of water vapor or of dilute sulphuric acid, this nitrosyl- 
sulphuric acid is at once decomposed, according to reaction (2), sul- 
phuric acid being formed, -and nitrous anhydride regenerated; or if 
sulphur dioxide and oxygen are concerned in the process, then reac- 
tion (3) occurs. This cycle of reactions repeats an indefinite number 
of times. But in the first lead chamber, where the temperature is 
rather high and an excess of water vapor is usually present, the fol- 
lowing secondary reactions probably occur to a greater or less extent : 

4) 2 SO 2 - (OH) (ONO) + SO 2 + 2 H 2 O = 3 H 2 SO 4 + 2 NO, 
this reaction being only momentary. 

* Ramsey and Cundall (J. Chem. Soc., 1885, 672) maintain that N 2 Os exists 
only as a liquid, and on heating, it decomposes into NO and NOs ; accepting this 
view, N2C>3, as such, cannot be present in the lead chambers, where the temperature 
is over 60 C. 

t Hurter (J. Soc. Chem. Ind., 1882, 49 and 83) supports the theory that nitro- 
gen peroxide (NO2) plays an important part in the process. 


Since there is usually an excess of oxygen present, however, the 
nitric oxide here formed is at once brought into action again, thus : 

5) 2 S0 2 + 2 NO + 3 O + H 2 O = 2 SO 2 (OH) (ONO). 

If there is a deficiency of oxygen, the nitric oxide is not returned 
to the process, but passes through the several chambers and, since 
it is not absorbed by the concentrated acid in the Gay-Lussac tower, 
it escapes into the atmosphere and is lost. 

The nitrogen oxides are derived from nitric acid, or by the action 
of sulphuric acid on sodium nitrate in the nitre pots. When nitric 
acid is used, it must be introduced in the form of vapor, or at least 
as a very fine spray, whereupon it reacts as follows : 

6) 2 SO 2 + 2 HNO 3 + H 2 O = 2 H 2 SO 4 + N 2 O 3 . 

Perhaps this reaction really occurs in two stages, thus : 

(a) SO 2 + HNO 3 = SO 2 (OH) (ONO) ; 

(6) 2 SO 2 (OH) (ONO) + H 2 O = 2 H 2 SO 4 + N 2 O 3 . 

The formula assigned to the nitrosylsulphuric acid may perhaps 

be written SO 2 , and the compound would then be called nitro- 

NO 2 

sulphonic acid. But in either case the existence of the substance is 
only transitory, it being broken up at once by the steam and sulphur 
dioxide present when the process is working properly. In case there 
is a deficiency of water vapor in the chambers, and especially if the 
temperature falls too low, the nitrosylsulphuric acid may separate 
as crystals, which deposit at various points on the walls, forming 
"chamber crystals." This is an undesirable accident, for when 
steam or water come in contact with them, they decompose into sul- 
phuric acid, nitric oxide, and nitrogen peroxide (N 2 O 4 ) : 

4 S0 2 (OH) - (ONO) + 2 H 2 = 4 H 2 SO 4 + N 2 O 4 + 2 NO. s 

Then the nitrogen peroxide unites with some of the water, 
N 2 4 + H 2 = HNO 2 + HNO 3 , 

forming nitrous and nitric acids directly on the walls, corroding the 
lead at the point where the cluster of crystals was attached. To 
prevent this separation of " chamber crystals " or retention of nitro- 


gen oxides in the sulphuric acid an excess of steam in the lead cham- 
bers is often preferred, although it dilutes the acid somewhat. 

Raschig,* after an extended study Of the process, maintains that 
nitrous acid (HNO 2 ) is present dissolved in the mist of sulphuric acid 
droplets filling the chamber. In the presence of air, water, and excess 
sulphuric acid, the sulphur dioxide and nitrous acid combine to form 
nitrosisulphonic acid, HO SO2 NOHO ; this then decomposes into 
sulphuric acid and nitric oxide. Finally the nitric oxide, reacting 
with water and air, is oxidized to nitrous acid. The cycle of re- 
actions thus becomes continuous, regenerating nitrous acid to react 
with new portions of sulphur dioxide, as follows : 

1) 2 HNO 2 + SO 2 = H 2 NSO 5 + NO. 

2) H 2 NSO 5 = H 2 SO 4 + NO. 

3) 2 NO + H 2 O + O = 2 HNO 2 . 

If, from any cause, the proportion of nitrous acid present falls below 
the quantity required by reaction (1), there is probably formed some 
nitrososulphonic acid, thus : 

4) HN0 2 + SO 2 = ONSO 3 H or (NO SO 2 OH). 

This may be one of the regular cycle of reactions of the process, but 
if an excess of nitrous acid is present, the nitrososulphonic acid passes 
over at once to nitrosisulphonic acid : 

5) HN0 2 + ONSO 3 H = H 2 NSO 5 + NO. 

But with an excess of water vapor present, the nitrososulphonic acid 
is hydrolized to form sulphuric acid and nitrous oxide : 

6) 2 (ONS0 3 H) + H 2 = 2 H 2 SO 4 + N 2 O. 

Since nitrous oxide is not absorbed in the Gay Lussac tower,f there is 
here a possible cause of the steady loss of nitrogen oxides (nitre) ob- 
served in every chamber system. 

The occasional formation of chamber crystals when water is 
deficient in the chambers is due to the formation of nitrosulphonic 
acid (HO . SO 2 . NO 2 ), by the oxidation of nitrosisulphonic acid by the 
nitric acid produced through reaction between excess oxygen and 
nitric oxide present in the chamber gases, thus : 

2 NO + 2 O = N 2 O 4 . 

N 2 4 + H 2 = HN0 2 + HN0 3 . 

H 2 NS0 5 + HNO 3 + HNO 2 = HNSO 5 + H 2 O + HNO 2 + NO 2 . 

* Zeitschr. angew. Chem., 1905, 1301 ; 1907, 701. 
t J. Soc. Chem. Ind., 1906, 149. 


The manufacture of chamber acid is shown in the diagram in 
Fig. 25. 

The acid may be made from brimstone, pyrites, blende, hydrogen 
sulphide, or the sulphur dioxide produced in metallurgical processes. 
Crude sulphur gives a pure acid free from arsenic, iron, copper, or 
zinc, and much smaller condensing chambers may be used for a given 
yield than when pyrites or blende is employed. Various types of 
brimstone burners are in use (see p. 58). 

Pyrites, or natural disulphide of iron (FeSfe), is a dense, hard 
mineral of crystalline structure and pale yellow color. Large 
deposits in the United States are in Virginia, at Mineral City, and 
at Charlemont in Massachusetts. Of the foreign deposits, those in 
Spain* are the most important. A pure pyrites contains 53.3 per 
cent of sulphur, but that commonly used for acid making carries 
from 43 to 48 per cent. It seldom pays to use an ore with less than 
35 per cent of sulphur, for it will not support its own combustion. 

The first proposal to use pyrites originated with an Englishman 
named Hill, who took out a patent for the process in 1818. But it 
was not until 1838, when the Sicilian government sold the monopoly 
of the sulphur export to a French firm which nearly trebled the 
price of crude brimstone, that pyrites began to find favor with acid 
makers. At the present time, because it is cheap and easily obtained, 
pyrites has almost completely replaced sulphur for acid making. 
The product from pyrites is usually contaminated with arsenic, and 
often with zinc, copper, and selenium. 

By the oxidation of pyrites in a suitable furnace, the sulphur 
is converted to dioxide, and iron oxide remains. The reaction may 
be written as follows : 

2 FeSs + 11 O = 4 SO 2 + Fe 2 O 3 . 

This is not exact, however, as some sulphur remains in the ore 
and some sulphur trioxide is formed. The proper regulation of the 
pyrites burners is one of the problems of the manufacturer. If the 
ore contains over 35 per cent of sulphur, the burning, once started, 
generates sufficient heat to maintain the combustion, and no fuel 
is necessary. But zinc sulphide and the " mattes " from metallurgi- 
cal processes must be heated by fuel. 

The complete burning of pyrites is difficult. With lump ore 
there is apt to be a kernel in the centre of the lump, from which 

* Spanish pyrites containing copper is much used in England and to some extent 
in this country, the burned cinder being afterwards treated to recover the copper. 



the sulphur is not burned out. If the temperature rises too high, the 
charge fuses together, forming clinkers or " scar," and choking 
the furnace. If too much air is admitted, the furnace cools below 
the temperature at which fresh pyrites will ignite, and the gases leav- 
ing the burner are so diluted that the desired reactions do not take 
place in the lead chambers. With " smalls " the tendency to fuse 
is more marked than with lump ore, and the fine ore packs together 
so densely that the air will not penetrate it, and unless it is constantly 
stirred only the surface is burned. (The lump ore is that which has 
been broken to about the size of a goose egg, the " smalls " constituting 
what will pass through a half -inch screen.) 

Pyrites burners are usually built in benches containing from three 
to thirty furnaces, in order that the supply of gas may not be broken 
while charging or cleaning one furnace. 

A burner for lump ore (Fig. 26) consists of a brick furnace, con- 
taining a grate formed of single loose iron bars (B, B) having a square 
section, and resting in 
grooves at each end. 
These bars may be 
turned parallel with 
their longitudinal axes, 
but have no lateral 
motion. They are so 
adjusted that their 
sides are at an angle 
of 45 to the vertical. After a charge is burned, the bars are given 
several quarter turns by means of a key, to allow the cinders on 
them to drop through into the ash pit. Air is admitted by dampers 
beneath the grate. When properly working, the cinders resting on 
the bars are nearly cold, the hottest part of the fire being eighteen 
inches above the grate. The furnaces are lined with fire-brick, 
and to prevent any access of air except through the dampers, the 
doors (D, K) for charging, cleaning, raking, etc., are made to fit 
closely, and are generally luted with clay. 

All the burners in one bench deliver their sulphur dioxide gas 
into a common, wide flue, or " dust box " (F), where any fine dust 
carried along by the gases may settle before they enter the Glover 
tower. This dust consists of unburned pyrites, arsenic, antimony or 
zinc oxides, iron oxide, etc. 

In one or more of the burners a cast-iron " nitre pot " may be 
set, in which nitrous gases are generated by the action of sulphuric 

FIG. 20. 



acid on sodium nitrate. Or the pots may be placed in a small chamber 
built into the flue (F), and heated by the waste heat of the burners. 
Sometimes, however, the pots are placed in separate furnaces. 

A lump burner of average size has a grate area of 15 to 25 square 
feet. The furnace is sometimes made slightly hopper-shaped in- 
side, so that it is larger at the level of the charging doors than at the 
grate bars. About 40 pounds of pyrites, containing 48 per cent of 
sulphur, are burned per square foot of grate area in 24 hours, a larger 
quantity of such high-grade ore being liable to cause fusion, unless 
great care is exercised. A larger quantity of poorer ore may be 
burned daily, without danger of fusion. 

A number of burners for fines have been invented, of which the 
Maletra burner (Fig. 27) is an early type. It consists of a series 

of shelves, about 5 by 8 feet in size, 
arranged in a tall furnace. The 
smalls, introduced through a hopper, 
fall on the top shelf and are spread 
out by rakes, introduced at the door 
(A) ; they can be dropped through 
an opening (B), upon the next shelf, 
to be again spread in a thin layer, 
and so on, each shelf being hotter 
than the preceding. The spent cin- 
ders are taken out at (C). In start- 
ing the furnace, the shelves are 
heated by burning brimstone or 
fuel, until the walls are hot enough 
to ignite the pyrites, which by their 
combustion evolve heat enough to continue the burning, as long as the 
furnace is properly regulated, and fresh ore supplied as needed. 

In Spence's furnace, fines are put into long muffles, externally 
heated. by waste heat from lump burners, or by generator gas, or a 
fire ; this furnace is used for roasting zinc blende, copper mattes, or 
concentrates, in which the sulphur is too low to burn, without external 
heat ; the sulphur dioxide gas is fairly concentrated. 

Raking shelf burners by hand is heavy labor and permits the en- 
trance of undue amounts of air; mechanical raking obviates this 
largely, and several appliances are in use. 

The Herreshoff burner * for fines has been largely introduced. 

FIG. 27. 

* This burner is a modified form of the McDougall furnace (see p. 552). 
eral Industry, Vol. VI, 236 ; XII, 267. J. Soc. Chem. Ind., 1899, 459. 




It is a steel cylinder about 11 feet in diameter, 9 to 10 feet high, and 
raised 3 feet from the ground on iron posts. It is lined with fire-brick 
and contains five slightly arched shelves, the top one having holes 
at the outer edge ; the next has a central opening ; the third at the 
outer edge, and so on. The cinders are discharged at the outer edge 
of the lowest shelf. A hollow cast-iron shaft, 14 inches in diameter, 
passing through the centre of the furnace, contains sockets into which 
the cast-iron rakes for moving the ore are fitted and locked by a 
simple lip catching in a notch. The shaft is steadied by a side bear- 
ing at the top of the furnace and is turned by a gear beneath the fur- 
nace bottom. From the upper end of the shaft a pipe extends into the 
open air ; at the bottom of the shaft, cold air is drawn in, and passing 
up through it and out by the pipe at the top, keeps the iron from 
becoming heated sufficiently for the sulphurous gases to act on the 
metal. As the shaft is rotated continuously, the rakes scrape the 
fines down from shelf to shelf, fresh 
ore being fed in, to maintain the 
combustion. Air for burning the 
pyrites is admitted through dampers 
near the bottom of the furnace, and 
the hot gases pass under and over 
each shelf as they ascend to the out- 
let at the top. The rakes may be 
easily replaced when broken, with 
only a few minutes' delay (compare 
Fig. 125, p. 597). 

The Wedge furnace (Fig. 28) is a 
recent development of the mechani- 
cal burner. It is much larger than 
the previously mentioned and has 
five to seven shelves. A steel central 
shaft, four or five feet in diameter and protected from the action of 
the hot gases by a fire-brick covering, carries the stirring arms set in 
cast-iron holders riveted to the shaft. This shaft is supported on, 
and rotated by a large gear about 12 feet in diameter, which runs 
on heavy rollers beneath the furnace, and is driven by a pinion and 
pulley. The hollow cast-iron arms are divided by a partition, which 
causes the cooling water or air to circulate through them, thus giving 
some control of the temperature on the separate hearths. Special 
shaped fire-brick are used in the shelves, which are flat on top, so the 
ploughs scrape the hearths clean. The top arch is used to dry the 

FIG. 28. 


material, which is then delivered at the centre on to a cast-iron feed 
plate, so arranged that the ore forms an air-tight lute, preventing any 
escape of sulphurous gases. The large size of the central shaft permits 
of workmen entering same to make repairs, without cooling down the 
furnace. These furnaces are built in several styles, some with muffle 
hearths, for desulphurizing, chloridizing, sulpha tizing, etc. (see p. 594). 
The Glover tower, used in nearly all sulphuric acid works, is 
placed next to the burners. Its functions are to set free the nitrogen 
oxides from the Gay-Lussac tower acid ; to cool the burner gases to 50 
or 60 C. before they enter the lead chambers ; to furnish part of the 
steam needed in the lead chambers ; and in many works to concen- 
trate the dilute acid from the lead chambers to a specific gravity of 
1.75. It also increases the yield of acid from a plant of given lead 
chamber capacity, for, in addition to that condensed in the chambers, 
some acid is formed in the tower itself. The tower (20 to 30 feet 
high and about 10 feet across) is made of sheet lead, joined as de- 
scribed below, and supported on a framework of timbers or steel. 
It is lined with acid-resisting brick or segments of Volvic lava,* laid 
without mortar, and is filled with quartz lumps, flint stones, or vitri- 
fied brick. At the top is an apparatus for distributing the acid, which 
is to run through the tower, f The burner gases enter at the bottom, 
and pass out at the top, by a pipe leading to the lead chambers. These 
form the most important part of a sulphuric acid plant, since in them 
the reactions involved in the formation of the acid take place. They 
are immense boxes, made by joining sheets of lead, and are supported 
from a strong timber or steel framework, by means of lugs or strips 
of lead attached to the outside of the sheets. The joints cannot be 
made with solder, but the edges of the sheets are fused together by 
means of an aero-hydrogen flame, and the process, called " lead burn- 
ing," is both difficult and slow. Steam or atomized water is intro- 
duced into the chambers to supply water vapor as needed. Each 
chamber is suspended above a large lead pan in such a way that the 
acid collecting in the pan forms a hydraulic seal for the lower edge of 
the lead chamber. These pans are 6 or 8 inches wider than the cham- 
ber, and have sides from 14 to 24 inches high. There is much dif- 
ference of opinion as to the best size and number of the lead cham- 
bers.J There are usually from 3 to 5, with a capacity of 140,000 to 

* An acid- and heat-resisting rock, found in the Puy de Dome* France. 

t The working -of the Glover tower is described in connection with the Gay- 
Lussac tower. 

J The so-called " tangential " chambers of Meyer consist of large cylindrical 
lead chambers, the inlet pipes placed tangentially on the sides and the outlet leading 


200,000 cubic feet in the system.* As a rule, the first chamber is the 
largest, and in it the greater part of the acid is formed. The individual 
chambers vary from 10,000 to 80,000 cubic feet (100 by 40 by 20 feet). 
In this climate they are enclosed in a building to avoid changes of 
temperature, which should not vary much from 50 to 65 C. in the 
first chamber, and 15 above that of the outside air in the last ; f 
and they are usually elevated, so that the acid may flow from them 
by gravity to the evaporating pans often placed on top of the pyrites 
burners ; and also that the bottoms may be better watched for leaks. 
To observe the working of each chamber, small lead dishes are fixed 
at various points on the inside of the chamber wall, and from these, 
pipes called " drips " lead to test glasses outside, where the density 
of the acid may be taken. A better method is to place the dish inside 
the chamber at a distance from the wall, supporting it above the 
level of the condensed acid, and connecting it by means of a pipe 
with a test glass outside. Glass panes are sometimes set at opposite 
points in the chamber walls, so that the color of the gases may be 
observed. In the first chamber the color is white and opaque, owing 
to the copious condensation of acid vapor, but in the succeeding 
chambers the color becomes more and more reddish, owing to the 
excess of nitrogen oxides. If the color becomes pale in the last cham- 
ber, there may be a deficiency of nitrous gases ; or too much or too 
little steam J ; or the draught may not be properly regulated, causing 
too much or too little oxygen to enter the chamber. The usual remedy 
is to introduce more nitre and then to locate the difficulty and grad- 
ually bring the system to its normal working condition. 

From the last lead chamber the gases pass to the Gay-Lussac 
tower, whose purpose is to recover the oxides of nitrogen. Sometimes 

from the centre of the bottom and becoming the inlet to the next ; thus the gases 
have a spiral movement which insures intimate mixing. 

Eng. Pat. No. 18376, 1898. Zeitschr. angew. Chem., 1899, 656: 1900, 739. 

* The usual American practice is to allow 16 to 22 cubic feet of chamber capac- 
ity per each pound of sulphur burned per 24 hours. In English practice for each 
pound of sulphur burned per day, from 22 to 29 cubic feet of chamber capacity is 

f Attempts have been made to operate at higher chamber temperatures, since 
a larger yield per unit of volume of chamber space is obtained, but these methods 
have generally failed, doubtless on account of the increased corrosion of the lead. 
Since a large excess of sulphur dioxide i~ present in the first chamber, the reduction 
of the nitrogen oxides is practically instantaneous and the corrosion is correspond- 
ingly low. Thus the maintenance of higher temperature there is feasible though 
not in the later chambers where the sulphur dioxide is nearly gone. 

J-The steam is derived from a boiler, or from the evaporation of water from the 
diluted tower acid in the Glover tower. 


two towers are used, the gases passing up through one and then to the 
bottom of the other, and up through this to the chimney. The tower 
is usually about 50 feet high, and 8 to 15 feet across. It is built of 
lead, supported on a frame, in much the same way as the Glover. It 
is lined with a double row of vitrified brick placed next to the lead 
walls, and inside of this is hard coke, or pottery rings, plates, saucers, 
or balls. At the top is a distributing apparatus to spread the acid 
evenly over the coke. The acid which flows down the Gay-Lussac 
tower is that which has been concentrated in the Glover tower to 
a density of from 60 to 62 Be. (about 1.750 sp. gr.). Acid of this 
strength absorbs the nitrous anhydride (N 2 O 3 ) and the nitrogen tetrox- 
ide (NO 2 or N 2 O4), but does not absorb nitric oxide (NO) or nitrous 
oxide (N 2 O). With normal working of the process, only that part 
of the nitrogen oxides is lost which is reduced to nitrous and nitric 
oxide. When an excess of oxygen is present, some of the nitric oxide 
is converted to nitrous anhydride, and thus saved. These nitrogen 
oxides are only absorbed when strong acid is run through the Gay- 
Lussac tower ; if the acid is of less than 1.50 sp. gr., it will not absorb 
them ; for best results it should be 1.75 sp. gr. The solution of nitrous 
gases in sulphuric acid, known in the works as " nitrous vitriol," 
is run into the Glover tower, where it is diluted with water, or cham- 
ber acid, till its specific gravity is about 1.6. As it passes down the 
tower, coming in contact with the hot sulphur dioxide from the burn- 
ers and steam from the lower part of the tower, the high temperature 
causes the dilute acid to give out its absorbed nitrous gases, which 
mix with the sulphur dioxide and pass back into the lead chambers. 
This process is called denitration of the tower acid. The heat in the 
lower part of the Glover tower evaporates a considerable portion of 
the water from the acid, thus concentrating it again to a strength 
sufficient for use in the Gay-Lussac, to which the required amount 
is returned, and the remainder is added to the acid which has been 
concentrated in the lead pans (p. 74). The hot burner gases are 
cooled by contact with the tower acid in the Glover tower to between 
50 and 60 C., the best temperature to work the first chamber. 

If the nitrogen oxides go to waste entirely, about 11 to 13 kilos 
of sodium nitrate must be used with each 100 kilos of sulphur burned. 
The recovery by means of the Glover and Gay-Lussac towers reduces 
the nitrate consumption to 4 kilos or less, per 100 kilos of sulphur, while 
a larger quantity of nitrous oxides is introduced into the chambers, 
causing the acid to form more rapidly and in greater quantities. 

Some manufacturers supply the nitrogen oxides in the form of 


liquid nitric acid, introduced into the chambers. This is easily regu- 
lated, admits no excess of air, and causes no loss of sulphur dioxide, 
such as may happen during the introduction of the " nitre." But care 
must be taken that the nitric acid does not run down the chamber 
sides, nor collect in the acid on the floor, for then the lead is rapidly 
corroded. Frequently the nitric acid is introduced into the Glover 
tower with the tower acid. The cost of the liquid nitric acid must be 
balanced against the advantages gained by its use. 

When sodium nitrate is decomposed by sulphuric acid in the nitre 
pots, the nitric acid vapor enters the bottom of the Glover tower with 
the sulphur dioxide. The vapors here coming in contact with steam 
begin to react at once, probably as follows : 

2 SO 2 + 2 HN0 3 + H 2 O = 2 H 2 SO 4 + N 2 O 3 ; 
or, 3 SO 2 + 2 HN0 3 + 2 H 2 O = 3 H 2 SO 4 + 2 NO. 

Thus the process of acid making begins in the Glover tower, and 
continues in the chambers according to the reactions given on p. 63. 

Sufficient sulphuric acid is used to form the acid sodium sulphate 
(NaHSO 4 ). This is liquid at the temperature prevailing and after 
the reaction is ended is easily run out through a tap in the bottom of 
the pot. On cooling, this acid sulphate solidifies, forming ".nitre 
cake" (p. 138). 

Compressed air is employed to force the 
concentrated acid from the Glover tower to 
the top of the Gay-Lussac, and the nitrous 
vitriol from the Gay-Lussac to the top of the 
Glover tower. The acid collects in a large 
oval vessel of cast-iron, called the acid egg FIQ 

(Fig. 29), and the compressed air from (B) 
forces it out through the pipe (A) to the Glover or Gay-Lussac tower. 

Kestner's acid elevator (Fig. 30) is much used ; a cast-iron lead- 
lined vessel (B) has a vertical pipe (T) in which a rod hangs free, 
extending from the air- valve case (D) to the float (X). Acid enters 
through the valve (M) and the pipe (A), lifting the float (X), which 
opens the valve in (D) by the rod in (T), admitting compressed air 
from the pipe (P) to (B) through (T). The air compressed in (B) 
closes the valve (M) and forces the acid out through the pipe (0) to 
the desired elevation. As the acid level in (B) falls, the float sinks 
until it closes the air valve (D), while acid again flows in through (A). 
The apparatus is automatic, simple, and occupies but little space. 
Modified forms are used for hydrochloric and other acids. The acid 



in the cistern supplying the apparatus must never reach a higher 
level than the line (FG). 

The "air-lift" pump (Fig. 31) is used to some extent to raise 
the acid to the top of the towers. A pipe (P) is sunk into the ground 
to a depth equal to the height to which the 
acid from (S) is to be raised ; the air from 
(R) is forced in near the bottom of the pipe, 
the pressure causing a rush of air up the 

FIG. 30. 

FIG. 31. 

pipe, carrying before it some of the acid, which is thus thrown out 
into (T) in " slugs," and not in a continuous stream. 

The acid condensed in the lead chambers varies from 1.5 to 1.62 
sp. gr. If more concentrated, it absorbs oxides of nitrogen present 
in the chambers, and attacks the lead. 

In the concentration of chamber acid it is first evaporated to 
1.70 sp. gr. (60 Be.) in shallow lead pans often heated by the waste 
heat from the pyrites burners. Since acid stronger than 1.70 attacks 
lead, '"oil of vitriol" is made in glass balloons, or in platinum or 
iron stills, or by direct heating in the Glover tower, Kessler apparatus, 
or dishes of porcelain or fused silica. 

Continuous acting concentrators, using porcelain or fused silica 
(Vitreosil, etc.) evaporating dishes set en cascade (Fig. 32) are con- 
siderably employed. Quartz dishes are bedded in asbestos rings on 



FIG. 32. 

fire-clay supports, so the flame strikes directly on the lower part of 
the vessel, but is not in contact with the acid, nor with the fumes 
from the evaporation. The acid over- 
flows from one vessel to the next, 
through a series of some 25 basins. 

Glass stills set in sand baths and 
heated by a fire are used somewhat, 
and yield a very pure, colorless, and strong acid ; but owing to break- 
age there is much loss and some danger. 

Platinum stills (Fig. 33) are shallow platinum dishes (S, S) cov- 
ered with a lead hood or bell (B), which is kept cool by a water jacket. 

The vapors condensing 
in this hood as a dilute 
acid do not fall back into 
the still, but collect in a 
narrow trough around 
the lower edge of the 
bell, and are usually re- 
turned to the lead pans. 
When the acid in the 
still has reached 1.835 
sp. gr. (66 Be.), it is 
drawn off through a platinum or lead cooling apparatus (C), as " oil 
of vitriol." The platinum stills are set directly over coke or coal 
fires on the grate (G), and are not allowed to cool except for repairs. 
Platinum stills may have a spiral partition in the pan which compels 
the dilute acid to flow a considerable distance over the hot still- 
bottom before it escapes through a tube from the central compart- 
ment. The rate of flow through the still determines the concentration 
of the acid. 

If the chamber acid contains nitrous vitriol, the platinum is often 
attacked. To prevent this, ammonium sulphate may be added to 
the acid during the concentration in the lead pans ; the nitrogen oxides 
are destroyed, thus : 

FIG. 33. 

N 2 O 3 + 

= 3 H 2 O + H 2 SO 4 + 4 N. 

Platinum alloyed with iridium is more resistant to the action of nitrous 
vitriol. A still invented by Herseus * consists of platinum lined with 
a layer of pure gold rolled with the platinum, and not electroplated. 
It resists the action of concentrated acid, but is attacked by nitrous 

* J. Soc. Chem. Ind., 1891, 460 ; 1892, 36. 



vitriol. The average loss of platinum in concentrating to " oil of 
vitriol " is about 1 gram per ton of acid produced. 

Cast-iron stills for highly concentrated acid are much used in 
modern work. These are usually shallow iron retorts, from 6 to 8 
feet long, by 2 to 4 feet wide, having a low cover provided with an 
outlet flue to carry off the vapors. The still is set so that it is en- 
tirely surrounded by the flame, thus preventing any condensation 
of dilute acid on the cover. Fins are often cast in the bottom to 
make the acid flow in a zigzag channel across the pan. Acid of 1.75 
sp. gr. or over has very little action on chilled cast-iron, and the stills 
stand from two to six months' constant use. The chamber acid is 
first concentrated to about 64 Be., in lead and platinum pans, or 
by running through the Glover tower, and then the hot acid enters 

Fia. 34.* 

the iron still and is brought up to the desired strength, ranging from 
93 to 98 per cent H2SO4. One type of cast-iron still setting is shown 
in Fig. 34 * ; the acid enters in a slow stream at (A), flows across the 
still and out at (B) into the vessel (C), where any sediment (sulphates, 
etc.) deposits. From (C) the concentrated acid flows into the coolers 
(E, E). 

Chamber acid is sometimes concentrated to 60 Be. in open lead 
pans heated by steam in lead coils ; this gives a clean product, but is 
not so economical as evaporation by waste heat from the burners. 
Over-surface evaporation (p. 4) in lead pans is occasionally practised, 
but yields a dark-colored acid. 

Kessler's acid-concentrating apparatus f (Fig. 35) is a combina- 
tion of over-surface heating with a tower evaporator. A chamber 
(G), built of siliceous materials enclosed in a lead case, is divided longi- 
tudinally by curtain partitions (P, P) ; over this chamber is a short 
tower (T), containing plates with overflow pipes (L) and porcelain 

* Trans. Am. Inst. Min. Eng., Vol. 16, 517. 
t J. Soc. Chem. Ind., 1892, 434 ; 1900, 246, 



FIG. 35. 

or fused quartz caps (J). The acid to be concentrated enters at 

(K), flows over the plates, and passes down by the pipes (L) from plate 

to plate, and finally to the chamber (G), where it lies about six inches 

deep on the floor; the curtain walls (P) just touch the surface of 

the acid. The hot gases 

from a coke fire enter at 

(E), pass under the lower 

edge of the curtain walls 

and into the channels 

leading to the tower. In 

passing under the walls 

(P) the hot gases bubble 

through the shallow layer 

of acid on the floor of (G), 

thus concentrating it ; the 

vapors and hot gases then pass up the tower, bubbling through the 

layers of dilute acid on the tower plates, and pass off through the 

hood and vapor pipe (V). 

When chamber acid is concentrated by running through the Glover 
tower, it is contaminated with iron from the flue dust of the burners. 
It is better to further concentrate such acid in cast-iron stills, since, 
when the density reaches 64 or 65 Be., a precipitate of ferric sul- 
phate forms, which may cake upon the platinum and cause it to crack. 
The acid intended for oil of vitriol is usually drawn from the lead 
pans, while that which has been through the Glover tower is fre- 
quently not further concentrated. 

To secure the intimate mixing of the gases essential in the lead 
chambers, Professor Lunge invented his plate tower,* a tall lead- 
lined tower divided into narrow chambers by transverse stoneware 
plates (Fig. 36) perforated by small holes, and so placed that the 
holes are not in line. By this arrangement the gases and liquids are 
brought into very close contact, and by placing such a tower between 
each pair of adjoining chambers, it is claimed that the chamber space 
for a given yield of acid can be much reduced. The plates are not 
practicable for the Glover tower, because the heat is liable to crack 
them, and the small holes become clogged with dust, but they may 
be used in the Gay-Lussac tower. 

* Zeitschrift fur angewandte Chemie, 1889, 385. J. Soc. Chem. Ind., 1889, 774. 
It may be noted here that these Lunge-Rohrmann "plate towers" have found 
much favor for condensing hydrochloric acid, but are said to obstruct the draught 
in sulphuric acid making. 




' ' : 


FIG. 36. 

The " pipe column " * invention of Gilchrist and Hacker and the 
towers of Hart and Baileyf carry out the same idea of mixing and 

cooling the gases more thoroughly. 
They consist of towers containing 
a number of small lead pipes set 
horizontally, and open to the air 
at each end. The gases, in pass- 
ing through the tower, impinge 
upon these tubes and are thus 
cooled and mixed, while air, pass- 
ing through the tubes, cools them 

also - 

The Barbier tower system,! in 
which the lead chambers are abol- 
ished and a series of towers substi- 
tuted, was carefully tested on a large 
scale in Italy, but the results were 
not satisfactory, probably due to 
excessive attack on the lead due to 

the high temperature (see footnote, p. 71). The advantages claimed for 
the system were : it occupies less ground and is cheaper to build than lead 
chambers ; it works at high temperature (90 C.), hence is less influenced 
by atmospheric changes, and is suitable for either hot or cold climates ; it 
gives a larger yield of acid per cubic metre of space than does the chamber 

While tower systems may be further developed in the future, the most 
promising substitute for the cumbersome and expensive lead chambers 
will probably be found in some of the " contact " processes (see below). 

To assist in the circulation and mixing of the gases in the cham- 
bers a fan of iron, hard lead, or earthenware is frequently placed in 
the inlet pipe, behind the Glover, or at the end of the system. This 
makes the working of the chambers uniform and independent of out- 
side temperature and wind. 

Atomized water instead of steam is often introduced into the 
lead chambers. This helps to abstract the heat liberated by the 
reactions, and increases the yield of acid per cubic foot of chamber 
space. The water must be in only the finest mist, made by directing 
a small jet, under high pressure, against a flat disk, or by using some 
type of spraying nozzle. 

Very concentrated acid may be made by artificially cooling oil 
of vitriol of 66.3 Be. considerably below C., when crystals of sul- 

* J. Soc. Chem. Ind., 1894, 1142 ; 1899, 459. 
$ Bui. Soc. Chim., 11, 726. 

t Ibid., 1903, 473. 

J. Soc. Chem. Ind., 1895, 698. 



phuric acid (monohydrate) separate, and are quickly freed from 
mother-liquor in a centrifugal machine. The crystals melt at 10 C., 
yielding an acid of 99.5 per cent H2SO4, with only a trace of water. 


The catalytic or contact processes had their origin chiefly in some 
experiments by Professor C. Winkler,* on the conversion of sulphur 
dioxide into sulphuric anhydride by the action of certain catalyzers. 
The fact of this conversion has long been known (Phillips, Eng. 
Pat., 1831), but no attempt to make practical use of it had been 
made. In 1878 Winkler patented a method for producing platinized 
asbestos to be used as a contact substance, and soon after other experi- 
menters began work along these lines. 

These processes attract manufacturers, since the plant occupies 
less ground area and does away with the costly lead chambers and 
the platinum-pan concentration ; all strengths of acids, from the 
weakest to the most concentrated monohydrate of 98.5 per cent 
H2SO4, and even fuming acid, can be produced in the same works, 
and with comparative ease. Further, no nitre, with the accompany- 
ing recovery process, is necessary. 

The raw materials are sulphur dioxide and oxygen from the air, 
to produce SOs. By solution of the sulphur trioxide in water, any 
concentration of acid can be made. 

The equation 2 SOz + Oz <^ 2 SOs shows a characteristic gas 

reaction. The equilibrium constant K p = ^ so ;! is given in the 

Pso 2 Vp 02 

following table; note that dissociation of the trioxide increases 
rapidly with the temperature : 

Degrees C. 
K, . . . 











In the absence of a catalyzer the rate of reaction is negligible below 
400 C. ; with finely divided platinum, combination may be detected 
at 200 C., and becomes rapid above 400 ; above 500 to 600 any 
surface is fairly active and burned pyrites cinder may be used. The 
reaction evolves 21.7 Cal., and unless the resulting rise of tempera- 
ture is controlled by dilution of the gases, and radiation of the heat, 
reversal of the reaction and destruction of the apparatus results. 

* Dingl. J., 1875, 296 ; 1877, 232 ; 1879, 384. 


This necessary temperature control is secured by enclosing the reaction 
chamber within the flue, in which the cold mixture of sulphur dioxide 
and air is passing to the catalyzer, thus cooling the contact mass 
and apparatus, and warming the mixed gases to the initial temperature. 
Or regulated quantities of the cold mixture are passed into the con- 
tact chamber at different points. By use of spongy platinum, the 
reaction may be carried on at 400 to 450 C., with nearly quantita- 
tive conversion; with less active accelerators, higher temperatures 
(500 C., or more) are required, and oxidation is less complete, neces- 
sitating recovery of the residual sulphur dioxide from the exit gases. 

The catalyzers most in use are spongy platinum and iron oxide 
from pyrites burners. The platinum mass may be platinized asbestos, 
or a sponge of metallic platinum disseminated through a porous mass 
of non-volatile soluble sulphates, oxides, or similar substance. 

The presence of flue dust, sulphur vapors, or of arsenic, phospho- 
rus, or mercury compounds in the mixed gases acts very injuriously 
upon the contact mass, soon rendering it inactive or causing rapid 
destruction of the apparatus. These substances must be entirely 
removed from the burner gases by cooling, scrubbing with water, 
injecting steam, or filtering. 

Cast-iron has proved unsuited for the construction of the appara- 
tus, since fuming acid makes it crack. The cause of this appears to be 
the formation of sulphurous acid in the pores of the iron, through the 
reduction of the acid by the action of the iron itself. Wrought iron 
seems to be passive to acid containing more than 27 per cent of sul- 
phuric anhydride and is well suited to the purpose. 

The contact process has entirely replaced the old dry distilla- 
tion of iron sulphate for fuming acid; it has also largely affected 
the manufacture of monohydrate and oil of vitriol. In this country, for 
making acid of 50 to 60 Be., the old chamber process appears to be 
economical, but in large plants, maintaining both processes, the expense 
of evaporating the chamber acid is avoided when making stronger 
grades, by adding contact sulphur trioxide to the weaker acid. 

The process of the Badische Anilin u. Soda-Fabrik* at Ludwigs^- 
hafen, Germany, was the first commercially successful one. In this, 
platinized asbestos is the contact material. The apparatus (Fig. 37) 
consists of several vertical iron tubes (R), containing perforated 
plates on which the platinized asbestos lies in thin layers, so that it 
does not offer too much resistance to the passage of the gases. The 
burner gases, cooled and purified, enter through (AA')> pass up the 

* Ber. deutsch. chem. Ges., 34 (1901), 4069. 



FIG. 37. 

space (S, S), between the tubes, thus cooling them, and thence through 
(0) and (F) to the chamber (D), from which they enter the tubes, 
pass down through the contact mass and 
out by (D') and (C). The tubes are first 
raised to the initial temperature by gas 
burners at (H), the combustion gases pass- 
ing out at (L) ; but once started, the heat 
of the reaction maintains the process. Thus 
the reaction heat is utilized to bring the 
mixture of SO 2 and air to the initial tem- 
perature, while the reaction products are 
cooled below the decomposition tempera- 

The Grillo-Schroeder process * employs 
platinized masses of soluble anhydrous salts, 
such as magnesium or sodium sulphate, as 
contact mass. This becomes inactive after 
a time, when the soluble salts are dissolved in water or acid and the 
platinum readily recovered. 

Hasenbach and Clemm propose to use the iron oxide residue 
from pyrites burning as contact material. This is not so effective 
as platinum, and the formation of sulphur trioxide is not near the 
theoretical amount, but the cheap material offers inducement for 
experiment. The pyrites cinders are introduced, still hot, into the 
contact chamber, which is a vertical shaft, and the burner gases 
require no purifying. Dust, arsenic, and other impurities are re- 
tained by the iron oxide in the lower part of the apparatus, and 
the anhydride is formed in the upper part. The cinder is removed 
periodically, as it becomes inactive. 

The difficulty of removing from the gas, dust and impurities (ar- 
senic) which poison the catalyzer is very great, especially when burn- 
ing pyrites. Frequently the gases are scrubbed in towers with strong 
sulphuric acid, or washed in spray chambers with a fine spray of acid ; 
then they are filtered through layers of pulverized coke, slag, or asbestos 
wool, before admitting them to the contact chamber. This difficulty 
of purifying the gases is a large factor in limiting the extension of the 
contact process. The operation is more sensitive and a higher grade of 
labor is required than for the chamber process, in which the elaborate 
treatment of the gases is unnecessary. Thus despite the expensive lead 
chambers and high cost of nitre, the latter holds its own for dilute acid. 
* J. Soc. Chem. Ind., 1899, 584; 1901, 579 4 



By absorbing the sulphur trioxide produced in the contact process 
in concentrated sulphuric acid, a brown, oily liquid is obtained, which 
fumes in the air, owing to the escape of some of the dissolved sul- 
phur oxides. Sulphur trioxide fume cannot be dissolved in dilute 
sulphuric acid, and hence concentrated acid must be used, which is 
later diluted to the desired strength. Fuming acid (" Nordhausen 
acid ") was formerly produced in Bohemia by the dry distillation of 
basic iron sulphates, obtained by weathering a kind of pyritiferous 
shale. When dried and heated in small retorts, decomposition ensues, 

Fe 2 (S0 4 ) 3 ' 2 FeSO 4 = 2 Fe 2 O 3 + 4 SO 3 + SO 2 . 

When absorbed in oil of vitriol, these vapors produced the fuming 


J. Soc. Chem. Ind., 1882 +. 

Progress in the Concentration of Oil of Vitriol. By W. H. Adams, Trans. 

Am. Inst. Min. Eng., 1887-1888. Vol. 16, p. 496. 
Mineral Industry. 1892 +. 

Schwefelsaurefabrication. Dr. K. W. Jurisch, Stuttgart, 1893. 
Die Gegenwartige Stand der Schwefelsaureindustrie. Gustav Rauter, 

Braunschweig, 1903. 

Sulphuric Acid and Alkali. Vol. I. 3d ed. G. Lunge, London, 1903. 
Ber. deutsch. chem. Gesell., 34 (1901), 4069. R. Knietsch. 
Zeitschr. angew. Chem., 1905 (18), 1253. (Chamber process.) 
Thermodynamik Technischer Gasreactionen. F. Haber, Berlin, 1905. 


The sources of salt are : 

1. Sea-water. 

2. Rock salt. 

3. Salt brines derived from springs, lakes, or wells. 

Atlantic sea-water, except near the mouths of large rivers, aver- 
ages about 3.4 per cent of solid matter, of which about 75 per cent 
is sodium chloride, the remainder consisting of chlorides, bromides, 
and sulphates of potassium, magnesium, calcium, lithium, etc., with 
minute amounts of other salts. 

The concentration of sea-water for salt is carried on to some 
extent in warm, dry countries by solar evaporation, the water usually 
being exposed in shallow tanks or ponds to the sun's rays. Sea- 
water is seldom evaporated over fire because of the cost of fuel. In 
Russia it is allowed to freeze over the surface, and the ice, which con- 
tains but little salt, is removed. This is repeated until the brine is 
sufficiently concentrated to make the evaporation over fire profitable. 
Salt made from sea-water ("sea-salt") is coarse and is usually damp, 
owing to the presence of some magnesium chloride, which, being a 
deliquescent substance, attracts moisture from the air. It is of less 
importance in this country than that made from other brines. 

Rock salt is found in many countries, and often very pure. In Eng- 
land, Austria, Germany, Spain, and Louisiana are large deposits, 
some so pure that it is only necessary to grind it for use, but in most 
cases it is contaminated with iron oxides, clay, sand, and other im- 
purities, which often necessitate its purification. In this country 
it is mined in New York, Kansas, California, Utah, and Louisiana. 
As it does not dissolve so readily as finely crystallized salt, it is 
preferred for many purposes, such as curing meat, preserving green 
hides, and feeding to live stock. 

The salt of principal interest in this country is derived from 
natural brines, found chiefly in New York, Michigan, Kansas, and 
Ohio, while West Virginia, Utah, Texas, and Pennsylvania produce 
lesser quantities. 

The New York deposits are near Syracuse and in the neighbor- 
hood of Warsaw and Batavia. The Onondaga (Syracuse) deposit 
has been known since the middle of the seventeenth century, but 
that at Warsaw, opened in 1883, is now the most important. The 



Michigan deposits are near Saginaw Bay and Manistee, a strong 
brine being obtained by boring. Large amounts of brine are evap- 
orated near Salina, Kansas. The Ohio and West Virginia deposits 
are in the valley of the Ohio River, near Pomeroy and Wheeling. 

Brines are obtained by bored wells, 8 inches in diameter, similar 
to those for petroleum (p. 336). The wells are lined with iron casings 
to exclude water from the over-lying strata. The brine as it comes 
from the well has some turbidity, due to clay or fine sand, together 
with minute bubbles of carbon dioxide, with which the brine is 
usually charged. Ferrous carbonate is also held in solution by the 
carbon dioxide, and on exposure to the air a yellowish red precipitate 
of ferric hydroxide separates. This is usually hastened by adding 
" milk of lime," or soda-ash, which also throws out some of the cal- 
cium and magnesium salts from the brine. 

" Solar salt " was formerly made in large amounts at Syracuse, 
and is yet produced at Great Salt Lake in Utah, and in California, 
from sea-water. The brine was exposed to the sun's rays in shallow 
wooden vats, from 6 to 8 inches deep. During the early part of the 
evaporation, crystals of gypsum, CaSO 4 2 H^O, separate in clusters, 
which are attached to the floor of the vat. After the gypsum is 
all separated, the brine is drawn into other vats, " salt-rooms," 
where evaporation causes the salt crystals to separate. These col- 
lect on the floor of the vat, and two or three times each season the 
salt is " harvested," i.e. raked up, freed from excess liquor in per- 
forated drainers, and removed to the store house. Wooden covers 
over the vats, which may be rolled back in fair weather, serve to keep 
out rain. In foreign countries " ricks " (p. 4) are used to concentrate 
brines, prior to evaporation for crystallization. 

Solar salt forms aggregates of the cubical crystals, which often 
take a " hopper " shape, and contain cavities in which small amounts 
of mother-liquor are retained, even after long draining. Since the 
liquors contain considerable amounts of calcium and magnesium 
chlorides, these contaminate the salt and cause it to become moist 
in damp weather. 

Strong brines, purified with milk of lime or soda-ash, are gen- 
erally concentrated by use of fuel, several types of evaporator being 
in use. 

The old " kettle process "* (Fig. 38), in which the evaporation was 
carried on in cast-iron kettles (A, A) about 4 feet in diameter, set in rows 
of 16 to 25 over a flue leading from the fire-box (G) to the chimney, has 

* After Merrill, Bui. N. Y. State Museum, III, No. 11. 



been generally abandoned. The brine was delivered through (P) to each 
kettle and the salt was raked out as it crystallized and drained in the 
basket (D), set over 
the kettle. A special- 
shaped " bittern pan " 
(B) was placed in the 
kettle at the start, and 
left until the salt began 
to crystallize, when it 
was lifted out, carrying 
much of the calcium 
and magnesium sul- 
phates, or " bittern," 
which separates first 
as the brine evaporates. 

FIG. 38. 

Sometimes the 
kettles are heated by 
steam jackets ; as all have the same steam pressure, the temperature 
is uniform, and only one quality of salt is produced. 

Salt is also made by the " pan process " (Fig. 39) * of direct 
evaporation over fire. Large wrought-iron pans (H, H), 24 feet 
wide, 100 feet long, and 12 inches deep, are used. These pans are 
divided into two sections by a loose partition, which allows the 
brine to flow slowly from the rear to the front section. A second 
smaller pan is set behind and slightly above the first, so that its 
contents may be syphoned into the front pan. Both are heated by 
flues from grates (G), but the rear one gets only the waste heat, 
before the gases pass into the chimney. The ends of each pan are 
made perpendicular to the bottom, but the sides are inclined, and 
sloping wooden platforms (F, F), called " drips," are joined to them; 
on these the salt is drained when removed from the pans. The 
brine is purified with " milk of lime," as in the kettle process. 

The pan process permits an easy control of the size of the grain. 
For the preparation of a very fine grained product, called " factory- 
filled salt," it is customary to add a small amount of sodium carbon- 
ate to the brine ; this decomposes the chlorides of calcium and mag- 
nesium and any excess of caustic lime from the " liming." Then a 
small quantity of butter, glue, or soft soap is added, and forms an 
insoluble calcium soap with the remaining traces of lime, and this is 
removed by skimming. 

For both the kettle and the pan process, coal dust is used as fuel. 

Strong brine boils at 105-109 C., and thus the heat in the kettle 
and pan process is sufficient to dehydrate any calcium sulphate in the 
* After Merrill, Bui. N. Y. State Museum, III, No. 11. 



salt; when dissolved in water, such products 
cause a slight milkiness, which disappears after 
a time, owing to the hydration of the calcium 
sulphate and its solution in the water. 

In Michigan and in western New York brine 
is evaporated in "grainers" (Fig. 40)*; these 
are long, shallow vats of wood or iron, contain- 
ing steam pipes (P, P), through which live or 
exhaust steam is passed. The pipes are about 

FIG. 40. 

4 inches in diameter and are hung about 6 inches 
above the floor of the " grainer," which is some 
20 inches deep. Once a day the salt is raked up 
and deposited on draining platforms over the 
grainers. The brine is purified before evapora- 
tion, as in the pan process, and is supplied to 
the grainer in just sufficient quantities to replace 
the water evaporated. When the mother-liquors 
become too highly charged with calcium and 
magnesium chlorides, they are drawn into special 
grainers, and a low grade of salt is made from 

Brine is frequently evaporated in continu- 
ous-acting vacuum pans, and a finely crystalline 
product, the best grade of table and dairy salt, 
results. It is separated from adhering mother- 
liquor by the centrifugal machine. 

Sometimes pure water is introduced into rock 
salt deposits through tube wells ; when saturated 
with salt, it is pumped to the surface and evap- 
orated. A much stronger brine than is found 
in nature is secured in this way. 

* After Merrill, Bui. N. Y. State Museum, III, No. 11. 

SALT 87 

In Michigan, West Virginia, Germany, and other places large 
quantities of bromine are recovered from the mother-liquors (also 
called " bittern ") from the salt industry. 

In Italy, Austria, and China the manufacture and sale of salt is 
a government monopoly. In France, Germany, and India salt used 
for seasoning food is subject to tax. When used for technical pur- 
poses, or in agriculture, the tax is very small. To prevent fraud, 
all German salt, not intended for table use, must be mixed with cer- 
tain substances to render it unfit for eating. Some of these adulter- 
ants are iron oxide, crude petroleum, coal dust, pyrolusite, carbolic 
acid, mineral acids, sodium sulphate or carbonate, alum, soot, etc. 


Die Industrie von Stassfurt und Leopoldshall. G. Krause, Cothen, 1877. 

Report on Manufacture of Chemical Products and Salt. W. L. Row- 
land, United States Census, 1880 ; Washington, 1884. 

Mineral Resources of the United States. (1882 +.) 

Chemische Industrie. 1883, 225. G. Lunge. 

Report of the State Geologist of New York, 1885, pp. 12-47. I. P. Bishop. 

Jour. Soc. Chem. Ind., 1888, 660. On the Tees Salt Industry. T. W. Stuart. 

Die Salz Industrie von Stassfurt. Dr. Precht, 1889. (Weicke, Stassfurt.) 

Bulletin of the New York State Museum, Vol. Ill, No. 11. Salt and 
Gypsum Industries of New York. F. J. H. Merrill, Albany, 1893. 

Forty-seventh Report of the State Museum of New York, pp. 205-257. 
The Livonia Salt Shaft. James Hall, 1894. 

Journal of the Society of Arts, 1894. Manufacture of Salt. F. Ward. 

Salt Deposits and Salt Industry in Ohio. J. A. Bownocker, Ohio Geol. 
Survey, Bull. 8, Vol. IX, 1906. 

Gewinnung und Reinigung des Kochsalzes. Carl Riemann, Halle, a. S., 

Louisiana Salt Mines. P. Wooten, Min. Eng. World, 1912, 401. 

The Salt Industry of Michigan. C. W. Cook. Mich. Geol. Survey, Pub. 
8, 1912. 


Hydrochloric or muriatic acid is generally made by the action of 
sulphuric acid on common salt. It is a by-product of the Leblanc 
soda process, and in the early years of the industry was allowed to 
escape into the air, as the demand for it was small. But the nuisance 
caused by the acid fumes in the neighborhood of the alkali works 
became so great, that in England a very stringent law was enacted 
forbidding the soda makers to allow more than 5 per cent of the gas 
to escape into the atmosphere. This made it necessary to absorb 
the acid fumes in water. The provisions of the present " Alkali Act " 
permit only 0.2 grain of hydrochloric acid per cubic foot of chimney 
gas to be discharged into the atmosphere. 

The Leblanc industry has declined in recent years, but there is an 
increased demand for hydrochloric acid, and at present this is one of 
the main products desired. Its chief use is for the generation of 
chlorine for the manufacture of bleaching powder; now nearly all 
soda makers also produce bleaching powder, and the profits derived 
from the latter have largely offset the decline in returns from soda- 
ash. Up to the present, no better method than the above has been 
devised for making this acid. The process may be represented by the 

equation : 

2 NaCl + H 2 SO 4 = Na 2 SO 4 + 2 HC1. 

But as actually carried out it takes place in two stages, according 
to the following reactions : 

1) NaCl + H 2 S0 4 = NaHS0 4 + HC1. 

2) NaHSO 4 + NaCl = Na 2 SO 4 + HC1. 

These reactions may be carried out by heating the mixture of 
salt and sulphuric acid either in an "open roaster," or in a muffle or 
" close roaster." These are both called " salt-cake furnaces." 

The open roaster (Fig. 41) consists of two parts, the cast-iron 
pan (A) and the reverberatory hearth (C). The salt and sulphuric 
acid (60 Be., sp. gr. 1.72) are put into the pan (A), and are moder- 
ately heated by a fire on the grate (E). The first reaction takes 
place at a comparatively low heat, and the hydrochloric acid vapors 
escape through the earthenware pipe (B). Then the fused mass 
of sodium acid sulphate and undecomposed salt is raked up on the 
reverberatory hearth (C), where it is exposed to the high temperature 




of the flame from (D). This completes the second reaction, and a 
pasty mass of normal sodium sulphate is formed. The hydrochloric 
acid vapors, set free 

during the reaction, gp 

mix with the furnace 
gases from (D), and 
escape through the 
pipe (F) to the ab- 
sorbing apparatus. 
The furnace gases 

dilute the acid vapors so much that a very concentrated solution 
of hydrochloric acid cannot be made with the open roaster ; however, 
it yields acid strong enough for use in Weldon's chlorine process 
(p. 117). Moreover, the soot and dust from the furnace at (D) con- 
taminate the acid, and may cause clogging in the passages and pipes 
of the absorption apparatus. The open roaster has the advantage 
over the close roaster that it yields more sodium sulphate with smaller 
consumption of fuel. The crude sodium sulphate, called " salt-cake," 
usually contains a little undecomposed salt and a slight excess of 
sulphuric acid. 

The muffle or " close roaster " is used very generally on the con- 
tinent of Europe, and yields a stronger and purer acid than the open 
roaster. The usual form is shown in Fig. 42. The pan (A) is built very 

much as in the open 
roaster, but is heated 
by the furnace gases 
from the grate (D). 
The acid vapors set 
free in the pan es- 
cape by the pipe 
(C) to the absorption apparatus. The muffle (B) is made of fire-clay 
or brick, and is heated by the flames from the grate (D). The 
mixture of acid sulphate and salt is raked from the pan (A) into 
the muffle (B), where it is heated to a red heat, and the acid vapor 
liberated passes through the pipe (E) to the absorption appara- 
tus. In this form of roaster the soot and dust from the grate are 
kept away from the acid vapor, and a concentrated acid vapor is 
obtained, which favors the formation of a concentrated solution of 
hydrochloric acid in the absorbers. But the muffles are expensive 
to build, yield a smaller output of salt-cake, and require more fuel 
than the open roaster. Moreover, they often crack, thus permit- 

Fia. 42. 



ting acid vapors to escape into the flues and chimney, causing loss 
and creating a nuisance. It is customary to maintain a slight pres- 
sure ("plus pressure") in the flues and chimney, so that if the 
muffle cracks, the flue gases force their way into it. This may cause 
a slight contamination of the acid, but no nuisance is created. 
Cheaper fuel may be used with these furnaces, but repairs are apt to 
be expensive. 

The pan (A) in both furnaces is about 10 feet in diameter, 7 inches 
thick at the centre, and 3 inches thick at the sides. After a charge 
is drawn, the pan is cooled somewhat before introducing another, 
for cold salt, coming in contact with the hot pan, might crack it. 
The sulphuric acid is generally heated to 
100 or 130 C. for the same reason. 

During the second reaction, the charge is 
constantly stirred with a " rabble," a large 
hoe-shaped tool, to prevent " crusting " or 
burning on to the hearth or retort. The 
stirring is very heavy work and the work- 
men are sometimes careless, and allow a 
crust to form, which may crack the muffle. 
Hence, many attempts have been made to 
construct mechanical stirrers. Of these, the 
Mactear furnace * is most successful, but the 
difficulty of protecting the driving mechanism 
from the acid fumes, and the cost of building 
and heavy up-keep charges, have caused gen- 
eral abandonment of mechanical furnaces. 

If salt-cake free from iron is desired, lead 
pans instead of cast-iron ones are used. But 
these are easily overheated or injured. 
The hydrochloric acid gas is absorbed in water, by passing through 
tall towers (Fig. 43) f filled with coke, over which water trickles ; or 
in a series of large earthenware Woulff bottles (bombonnes or tourills, 
Fig. 44 {), with an absorption tower at the end to catch acid gas which 
may pass through the bottles. These are set en cascade and the side 
tubulatures joined so that a stream of dilute acid from the tower 
flows through them in opposite direction to the movement of the gas. 

* Chemische Industrie, 1881, 253. J. Soc. Chem. Ind., 1885, 534. 
t After Lunge. 

J Metal. Chem. Eng., 1911, 611. 

That is, on a series of steps, so that each stands slightly lower than the one 

FIG. 43. 


The standard absorber is difficult to cool externally with water and 
presents relatively a small liquid surface to the gas. A modified form 

(Fig. 45) * is claimed to be better, as 
it affords greater liquid surface exposure, 
and can be readily water-cooled. 

The Lunge-Rohrmann plate tower 

FIG. 44. 

FIG. 45. 

(p. 77) has been tried with some success as a substitute for the coke 
tower and bombonnes, for hydrochloric acid absorption. 

The condensation of hydrochloric acid vapors is not so simple a 
process as it at first appears. The gases coming from the roasters 
are very hot, and must be cooled before they can be absorbed to 
form a strong acid. Moreover, with open roasters, there is a large 
amount of inert gas present (nitrogen and carbon dioxide from the 
fire) which dilutes the acid vapors. Then, too, the vapors are not 
set free regularly in any roaster, there being a rapid evolution during 
the progress of the first reaction, and a much slower liberation during 
the second. This may cause a temporary rush of vapors through the 
apparatus, so that they cannot be properly taken up by the water. 

The ordinary muriatic acid of trade is an aqueous solution of the 
acid vapor, having a specific gravity of about 1.20 and containing 
about 40 per cent by weight of dry hydrochloric acid vapor. It is 
impure, containing sulphuric acid, chlorine, iron chloride, arsenic, 
and, generally, lead and calcium chlorides. Its yellow color is partly 
due to organic matter, and sometimes to iron and free chlorine. To 
remove arsenic and sulphuric acid, the acid is diluted to 1.12 sp. gr., 
and barium sulphide is added; a pure hydrochloric acid vapor is 
then driven out by distillation and absorbed in pure water. Or a solu- 
tion of stannous chloride in concentrated hydrochloric acid is added 
to the crude acid, which latter must have a strength of at least 1.15 
sp. gr. A brown precipitate of arsenic with some tin separates and 
is removed by decantation.f Sulphuric acid alone is removed by 

* Metal. Chem. Eng., 1911, 611. 

t 3 SnCh + 6 HC1+ As 2 O 3 = As 2 + 3 H 2 O + 3 SnCh. 

2 AsCls + 3 SnClz = As 2 + 3 SnCh. 
This leaves stannic chloride in the acid. 


adding barium chloride and redistilling. To remove chlorine, the 
crude acid is digested with strips of copper for some hours. This 
precipitates arsenic, and the chlorine combines with the copper. The 
acid is then redistilled. 

Attempts to recover hydrochloric acid from the waste liquors of 
the ammonia soda process (p. 122) have not proved very successful. 
The magnesium chloride mother-liquors from the potash salts of 
Stassfurt (p. 160) may be decomposed by distillation with steam, 
and a dilute hydrochloric acid obtained. 

MgCl 2 + H 2 O = 2 HC1 + MgO. 

But this has not proved a commercial success. 

The Hargreaves and Robinson process for the direct production 
of hydrochloric acid and sodium sulphate from salt, sulphur dioxide, 
water, and oxgyen is of some importance. The damp salt is pressed 
into blocks and dried ; it is then charged into vertical cast-iron retorts, 
a number of which are connected in a series. These are heated from 
without; the temperature of the reaction is from 400 to 550 C. 
The sulphur dioxide, steam, and air are made to pass through all 
the retorts in succession, the hydrochloric acid being carried along 
with them. A slight excess of sulphur dioxide and steam is used to 
prevent the mutual reaction between the hydrochloric acid vapor 
and the oxygen, by which chlorine is set free. The decomposition 
being slow, the gases must be kept in contact with the salt for a 
considerable length of time ; a cylinder containing 40 tons of material 
requiring from 15 to 20 days' continuous action to secure complete 

The process is an uninterrupted one ; for as soon as no more sul- 
phur dioxide is absorbed in a given cylinder, it is cut out from the 
series, the sodium sulphate removed, a new charge of salt blocks 
introduced, and the cylinder made the final one of the series; so 
that newly charged salt is exposed to the most nearly exhausted 
sulphur fumes. The reaction representing the process appears quite 

simple : 

2 NaCl + SOa + H 2 O + O = Na 2 SO 4 + 2 HC1. 

But the mechanical difficulties encountered in working it were great, 
and only recently has the process met with any marked success. 

Sodium sulphate or salt-cake is largely used in the production of 
soda by the Leblanc process, for glass making, for ultramarine, in 
dyeing and coloring, and to some extent in medicine. For some kinds 
of glass the salt-cake must be free from iron, and consequently it is 


made in lead pans. Or the sulphate may be purified from iron and 
excess of acid by dissolving it in hot water, adding " milk of lime," 
and stirring into it a solution of bleaching powder. The iron is pre- 
cipitated as hydroxide and settles on standing. By evaporation, crys- 
tals of Glauber's salt (Na 2 SO 4 10 H 2 O) are obtained. But generally 
the purified solution is rapidly evaporated to dryness, and the product 
is calcined to remove all the water. 


Berichte iiber die Entwickelung der Chemischen Industrie, u.s.w. A. W. 

Hofmann, Braunschweig, 1877. (Vieweg.) 
Darstellung von Chlor und Salzsaure, unabhangig von der Leblanc Soda 

Industrie. Dr. N. Caro, Berlin, 1893. (Oppenheim.) 
Sulphuric Acid and Alkali. 3d ed., Vol. II. G. Lunge, London, 1909. 

(Gurney and Jackson.) 
Die Fabrikation von Sulfat und Salzsaure. Theo. Meyer, Halle, a. S., 




Nearly all the soda of trade was formerly obtained from certain 
natural deposits of the so-called " sesquicarbonate," or from the ashes 
of sea plants. But towards the end of the last century, the sup- 
ply from these sources became insufficient to meet the increasing 
demands. About 1775 the French Academy of Science offered a 
large prize for a method of making soda from salt. Among other 
processes submitted was one by Nicolas Leblanc, which seemed prom- 
ising, and being granted a patent in 1791, he began manufacturing 
on a commercial scale. But in the French Revolution his factory 
was seized, the patent declared public property, and no indemnity 
was paid to him. Having lost all his property, he finally committed 

Leblanc's process was so perfect and complete that very slight 
changes, and those only in minor details, have been made up to the 
present. It has been in use for more than a century, and although 
seriously threatened by newer processes, it still produces a large 
part of the world's supply of soda. Owing to the fact that it pro- 
duces hydrochloric acid and bleaching powder as by-products, it has 
been able to survive competition, although its condition is becom- 
ing more desperate every year. Its chief rival is the ammonia or 
Solvay process. Within a few years many electrolytic methods for 
caustic soda have appeared, and the extensive production of bleach- 
ing material by any of these processes will sweep away about the 
only source of profit left to the Leblanc manufacturer. It is not 
probable that this change will come immediately, although several 
electrolytic processes have proved fairly successful on a large scale; 
but the decline of the Leblanc process is generally regarded as inevi- 
table, and inventors have, for the most part, abandoned further 
attempts to improve it. 

The reactions of the Leblanc process are generally expressed as 
follows : 

1) 2 NaCl + H 2 S0 4 = Na 2 SO 4 + 2 HC1. 

2) Na 2 SO 4 + 2 C = Na 2 S + 2 CO 2 . 

3) Na 2 S + CaCO 3 = Na 2 CO 3 + CaS. 

4) CaCO 3 + C = CaO + 2 CO. 




But these equations * do not represent all the reactions which take 
place during the process, for a number of other substances are formed. 
The first equation represents the preparation of sodium sulphate and 
hydrochloric acid (p. 88). The second and third reactions are real- 
ized in one operation. The fourth has no direct relation to the process, 
as the formation of carbonic oxide does not become marked until all 
the salt-cake has been decomposed. This serves to indicate the end 
of the process, and aids in the formation of a porous product. 

The salt-cake should be friable and porous, containing very little 
free sulphuric acid, and no undecomposed chloride. The carbon 
is supplied in the form of powdered coal, which should contain very 
little ash-forming impurity. A little pyrite does no harm, but the 
coal should be as free as possible from nitrogen, in order to prevent 
the formation of cyanides and cyanates. Calcium carbonate in the 
form of pure limestone or chalk, crushed to the size of a small pea, 
is mixed with the crushed salt-cake and coal in order to carry out 
the third reaction. If the limestone contains magnesia or silica, 
there is a consequent loss as insoluble residue. Usually 100 pounds 
of salt-cake, 100 pounds of limestone, and 50 pounds of coal dust 
form a charge. This is an excess of limestone, the purpose of which 
is explained below. 

The reactions are carried out in a " black-ash " or " balling fur- 
nace," which may be worked either by hand or mechanically. The 
hand-worked furnace is a long reverberatory (Fig. 46). The charge 

FIG. 46. 

is introduced on the platform (A) nearest the flue, where the 
heat is not high. When well heated, it is raked on to the front 
platform (B), which is a few inches lower than (A). Here the tem- 

* Lunge (Sulph. Acid and Alkali, Vol. II, 460 et seq.) regards the theory of 
Scheurer-Kestner (Comptes rendus, 57, 1013, and 58, 501) as correct, viz. that 
the reactions are : 

5 Na 2 SO4 + 10 C =5 Na 2 S + 10 CO*. 

5 Na 2 S + 5 CaCO 3 = 5 Na 2 CO 3 + 5 CaS. 
2 CaCOs + 2 C = 2 CaO + 4 CO. 

While these equations closely represent the net result, the first and last re- 
actions each yield mixtures of CO 2 and CO, depending on the temperature and 
equilibrium conditions. 



perature is high, usually about 1000 C., and the surface of the mass 
soon begins to fuse. It is then raked over, thoroughly exposing it 
to the direct heat until it becomes a thick, pasty mass, from which 
carbon dioxide is escaping freely. After the salt-cake is all decom- 
posed, the charge begins to stiffen, and the evolution of carbon mon- 
oxide is shown by the appearance of jets of blue flame, known to the 
workmen as " candles." The charge is then raked together into a 
" ball," which is drawn out of the furnace into an iron barrow. The 
evolution of carbon monoxide continues for a few minutes after 
the " ball " is removed, and the bubbles escaping from the pasty mass 
cause it to become porous. The formation of this gas is due to the 
action between the coal and the excess of limestone according to 
reaction (4). The caustic lime formed here slakes during the lixiv- 
iation of the black-ash (p. 97), and swells, thus disintegrating the 

Although the heavy tools are suspended by chains, their operation 
is still so difficult, and the temperature is so high, that a man cannot 
handle much more than 300 pounds at one time. In order to work 

larger charges, without the expensive hand labor, revolving black- 
ash furnaces (Fig. 47) are much used. These are similar to the re- 
volving furnaces described on page 21 ; the flame from the furnace (A) 
passes through the cylinder (B). The charge is introduced through 
the manhole (P), and the finished product discharged through the 
same opening, into the wagon, at the end of the operation. The 
cylinder is about 16 feet long by 10 feet in diameter, and is revolved 
by a gear (E) connected with an engine. Projections are fixed in the 
lining to help mix the contents. The charge is usually about two 
tons of salt-cake, with proportionate amounts of coal and limestone. 
It is customary to introduce only the limestone and a part of the coal 
at first, and to rotate the cylinder until some caustic lime is formed ; 
then the remainder of the coal, together with the salt-cake, is intro- 
duced, and the rotation continued until the reactions are completed. 
The speed varies from one revolution in three or four minutes, at first, 


to four or five revolutions per minute during the last part of the 

The hot gases from the black-ash furnace, whether hand-worked 
or mechanical, pass through the dust box (N), and then through the 
long flue over the pan (J, J) on their way to the chimney (D). In 
this shallow pan, the liquor obtained by lixiviating the black-ash is 
evaporated. When crystallized, the salts are removed through the 
small doors (J). 

Black-ash is a brownish black or dark gray substance of a pumice- 
like texture, containing about 45 per cent sodium carbonate, 30 per 
cent calcium sulphide, 10 per cent caustic lime, and from 10 to 12 
per cent of other impurities, sulphate, silicate, aluminate, and 
chloride of sodium, calcium carbonate, coal, and iron oxide, with 
traces of cyanides and of sulphides of sodium. 

The next stage in the process is the lixiviation of the black-ash. 
This presents some difficulties : if the black-ash is put directly into 
cold water, it often agglomerates in hard lumps, which dissolve ex- 
ceedingly slowly ; the free lime present forms calcium hydroxide, 
which reacts with the sodium carbonate solution, forming some caustic 
soda ; the solution of sodium carbonate, especially if hot and dilute, 
reacts on any calcium sulphide present, forming some sodium sulphide ; 
moreover, moist calcium sulphide oxidizes rapidly to sulphate in the 
air, and this reacts with the sodium carbonate. Hence the 'lixivia- 
tion must be done as rapidly as possible, at a low temperature, and 
without exposing the wet black-ash to the air. 

Shank's process gives the most satisfactory results. The lixivia- 
tion is carried on in a series of tanks, each having a false bottom per- 
forated with small holes. Because of its density, the solution of 
sodium carbonate sinks, and passing through these perforations, is 
drawn off by means of a pipe which delivers it at the top of the next 
tank. There must always be sufficient liquor in each tank to keep 
the black-ash entirely submerged. The process is continuous, suffi- 
cient fresh water being admitted to the nearly exhausted ash to give 
an unbroken flow of strong liquor (above 45 Tw.) from the last tank 
of the series. When the liquor from the last tank falls to 45 Tw., it 
is turned into a tank which has just been filled with new ash. The 
exhausted ash is washed until the wash water has a density of only 
1 Tw. Then the residue of calcium sulphide and hydroxide, coal, 
ashes, and other insoluble matter, which constitutes the " tank waste," 
is dumped and the tank refilled with black-ash arid made the last of 
the series, to receive the strong liquors from the preceding tank. 


Since the black-ash contains caustic lime, sufficient heat is gen- 
erated by its slaking during the lixiviation to warm the concentrated 
liquor to about 50 C., which is the best temperature for complete ex- 
traction. The temperature of the dilute lye from the first tank of 
the series is not allowed to rise above 38 C., in order to prevent the 
above-mentioned interaction between the calcium sulphide and the 
sodium carbonate. 

Good tank liquor has approximately the following composition : 

Na 2 CO 3 (+NaOH) 23.60* 

NaCl 50 

Na^S 13 

NajSaO, 30 

Na*SO 4 23 

Na 2 SiO 3 ) 

NaCN Traces. 


FeS (in solution) J 

The lye obtained by the lixiviation has a specific gravity of about 
1.25, and is muddy from suspended impurities. It is purified by 
settling and then pumped to the top of the " carbonating towers," 
which are filled with pebbles or coke, or have numerous chains or 
wire ropes suspended from the top and weighted at the lower ends. 
The tank liquor trickles over the porous material or chains, and 
comes into intimate contact with a strong current of carbon dioxide f 
entering at the bottom and passing up through the tower. 

The carbon dioxide and oxygen which pass through the tower 
convert the caustic soda to carbonate, decompose the ferro-sodium 
sulphide (solution of ferrous sulphide in sodium sulphide), convert- 
ing the sodium sulphide into bicarbonate, and precipitating the iron, 
together with any silica and alumina which may be present. 

The reactions involved were supposed to be the following : 

1) 2 NaOH + CO 2 = Na 2 CO 3 + H 2 O. 

2) Na 2 S + C0 2 + H 2 = NaHCO 3 + NaSH. 

3) NaHCO 3 + NaSH = Na 2 CO 3 + H 2 S. 

Reactions (2) and (3) are incomplete, as hydrogen sulphide is of 
practically the same acid strength as carbonic acid ; only when 

* Mohr, Analysis of Soda-ash from Stolberg (Lunge, Sulphuric Acid and Alkali, 
Vol. II). 

t This is derived from the gases of the black-ash furnace, which also contain 
some oxygen. Or it is obtained from the gases from lime kilns, which are much 
richer in carbon dioxide and introduce less flue dust into the product. 


enough CO 2 is used to convert all the carbonate to bicarbonate, or 
when a large excess is used at fairly high temperature, can all the sul- 
phide be decomposed. By adding zinc hydroxide the sulphide may 
be precipitated : 

Na 2 S + Zn(OH) 2 = ZnS + 2 NaOH. 

Or by blowing air through the tank liquor the sulphide is converted 
to thiosulphate : 

NmS + 2 O 2 + H 2 O = 2 NaOH + NaSA. 

Pauli's process of purifying tank liquor by adding " Weldon mud " 
and blowing in air and steam is more effective. Thus the sulphide is 
oxidized, and ferric oxide, alumina, and silica precipitate in the sludge. 
Assuming " Weldon mud " to be essentially manganese dioxide, the 
following reactions take place : 

2 Na^S + 4 Mn0 2 + 5 H 2 O = 2 NaOH + NasSA + Mn(OH) 2 . 
Mn(OH) 2 + 2 2 = 4 MnO 2 + 4 H 2 O. 

The manganese oxide thus recovered is used repeatedly, until it be- 
comes much contaminated with ferric oxide, alumina, silica, etc. 

After settling, the purified and carbonated tank liquor is drawn 
into the evaporating pans, usually large shallow iron tanks, and heated 
by surface contact with the waste gases from the black-ash furnace. 
Sometimes deep pans, heated from below, are used, since surface 
evaporation gives a product contaminated with dust from the furnace. 
The liquor is evaporated directly to dry ness, and the " black salt " 
(chiefly monohydrated sodium carbonate, Na 2 COs H 2 O) is calcined 
by heating it to a red heat. Sometimes sawdust is mixed with the 
uncarbonated liquor before evaporation, and then on calcining, the 
soda-ash is carbonated by the carbonaceous matter from the wood ; 
but the charge is liable to cake in this method. The caustic soda and 
sodium sulphide of the tank liquor are thus converted to sodium car- 
bonate, and, after the sawdust is burned out, the ash becomes white 
or light brown. 

Or the liquor is evaporated till a crystalline mass separates ; then 
the mother-liquor (" red liquor ") is drawn off, and the black salt is 
raked out of the pan. Much care is necessary to prevent the forma- 
tion of a crust or the burning on of the precipitated carbonate. 

In large works, a semi-cylindrical evaporating pan is used, 
provided with mechanical scrapers, to prevent the black salt from 
adhering to the pan. An excellent form of this apparatus is 



Thelen's pan (Fig. 48). In this, the scrapers (R, R) move the salts 
towards the end of the pan as they deposit, and a scoop lifts them to 
the draining apron. The beam (B), carrying the frame from which 
the scrapers are suspended, is rotated by the gear (J). 

For a very light-colored product, the crude soda-ash is dissolved 
in water, and a little bleaching powder solution added ; the precipi- 
tated iron and other impurities settle out, and the clear solution is 
evaporated until a thick mass of crystals separates, when the mother- 
liquor is drawn off to remove any soluble impurities. The mono- 
hydrated salt remaining is then calcined without the addition of car- 
bonaceous matter, to remove its crystal water, and the product is 
called " white alkali " or " refined alkali." A little sodium chloride 
is formed by the addition of the bleaching powder, so that refined 
alkali is not quite so strong as soda-ash. It is chiefly used for glass 
making and other purposes where iron and sulphides would be detri- 
mental. Good Leblanc soda is nearly white or pale yellow, and should 
contain but few black specks. It usually contains a little caustic 
soda, a trace of sulphides and sulphites, some chloride and sulphate, 
and not over 1 per cent of insoluble matter. It should be finely 
ground before packing. 

Soda crystals or sal-soda (Na 2 CO 3 10 H 2 O) is made by dissolving 
soda-ash in warm water, allowing the hot solution to stand quietly 
until all sediment deposits, and drawing off the clarified liquor into 
crystallizing tanks, where is it cooled to the atmospheric temperature. 
Large crystals of sal-soda, very nearly pure, are deposited. They 
contain over 60 per cent of water, and are thus very bulky and not 
economical to ship ; but they are still preferred to soda-ash for some 
manufacturing purposes and for household uses. They are some- 
times used for making sodium bicarbonate, by exposing them on a 
grating to an atmosphere of carbon dioxide : 

Na 2 CO 3 10 H 2 O + CO 2 = 2 NaHCO 3 + 9 H 2 O. 


The water resulting from the "reaction drips through, leaving 
the bicarbonate on the grating. 


Caustic soda is made from soda-ash, or from the " tank liquors " 
directly, by adding calcium hydroxide (milk of lime) to the solution : 

Na 2 CO 3 + Ca(OH) 2 = CaCO 3 + 2 NaOH. 

When caustic soda is the ultimate product, it is generally custom- 
ary to use this lime mud (CaCO 3 ) instead of limestone, in the charge 
for the black-ash furnace, for the formation of caustic in the tank 
liquor is then of course not objectionable. 

The tank liquor must not have a density of over 20 Tw. (1.10 
sp. gr.), or the reaction will be incomplete.* Consequently it is 
diluted with the wash waters from the lime mud of a previous opera- 
tion. The liquor is then heated to boiling, and run into large iron 
tanks, where the " milk of lime " is added, and the mixture well 
stirred. Air or steam is usually blown into the liquor to assist in the 
mixing. The air, especially when aided by the addition of " Weldon 
mud " (p. 118), converts the sodium sulphide to sodium thiosulphate 
and sulphate : 

2 NajS + H 2 O + 4 O = 2 NaOH + 

The thiosulphate is afterwards destroyed by oxidizing it to the 

* This reaction may be written in the ionic form, thus : 
COr- + Ca(OH) 2 2 OH-+ CaCO 3 . 

Equilibrium will therefore correspond to the equation, - - - - = Const., where 

(OH-) 2 

the parenthesis indicates that the concentration of the ion enclosed is meant. It is 
desired to produce a high hydroxyl concentration relative to the carbonate, but this 
is attained only at low total concentration, since the influence of the hydroxyl ion 
tending to reverse the reaction is proportional to the square of its concentration while 
the carbonate is only in the first power. In fact in dilute solutions the reaction goes 
almost quantitatively to the right, but at 20 Tw. the conversion is only about 85 
per cent, and at higher concentrations it is even less. These facts have been used as 
text-book illustration of the industrial application of the law of mass action, but it is 
not the incompleteness of the conversion which is the reason for the technical use of 
dilute solutions, since any unconverted carbonate is recovered on concentrating. 
It is rather the fact that the precipitated carbonate requires, in any case, large 
amounts of water for washing, which must be evaporated. Since this water, if 
used in the original batch, gives the advantage of high causticization, with no dis- 
advantages, it is of course employed. 


In large plants, the dilute caustic solution, after settling, is con- 
centrated in multiple-effect systems to a density of 60 Tw., at which 
point the other salts, such as sodium carbonate or chloride, still dis- 
solved in the liquor, begin to crystallize ; the liquor is then run into 
cast-iron kettles (" pots "), which are heated by direct fire, and the 
last of the water evaporated, when caustic soda remains as a fused 
mass. The salts separated during the evaporation are raked out. 
Some nitre may be added, or air blown in, to complete the oxidation 
of any thiosulphate to normal sulphate, which remains in the caustic, 
reducing its strength. In small works the dilute liquor is evaporated 
directly in the iron pots. For strong caustic zinc oxide is often added 
to remove the sulphide from the tank liquors. The precipitated zinc 
sulphide is removed by settling and calcined to reconvert it into 

The fused caustic soda is run directly into the sheet-iron drums 
in which it is sold. These are sealed as soon as cold, to prevent the 
absorption of moisture by the caustic. 

Loewig's process * for caustic soda depends on the formation of 
sodium ferrate (Na 2 Fe 2 O 4 ), which is then decomposed with water. 
The soda liquors are mixed with ferric oxide, and the mass evaporated 
to dryness and calcined at a bright red heat, usually in a revolving 
furnace. By the calcination, a reaction between the sodium carbonate 
and the iron oxide is brought about, carbon dioxide escaping and 
sodium ferrate remaining in the furnace. The mass is washed with 
cold water until all soluble matter is removed ; then water at 90 C. 
is run over the sodium ferrate, by which it is decomposed, caustic 
soda formed, and iron oxide regenerated ; the last is returned to the 
calcining process. The ferric oxide used is a natural iron ore, very 
clean and free from silica or other impurities ; that made by calcining 
a precipitated ferric hydroxide is not well adapted to the process, as 
it gives a product difficult to lixiviate. The density of the resulting 
solution of caustic is about 62 Tw. (1.31 sp. gr.), and so less evapora- 
tion is necessary than in the lime process, where the density is only 
15 or 20 Tw. Moreover, there are no other salts present, such as 
sulphate, thiosulphate, sulphide, or chloride, and the product is purer 
than that yielded by the lime process. But Loewig's process is not 
so well adapted to use with the Leblanc soda-ash, because the tank 
liquors must be evaporated to dryness before calcining the ferric oxide 
and sodium carbonate mixture, and the sodium carbonate must be 

* German Patent, No. 1650, Dec. 21, 1877. J. Soc. Chem. Ind., 1887, 438. 
Konrad W. Jurisch, Die Fabrikation von Schwefelsaurer Thonerde, p. 13. 



pure. The process may be advantageously used with ammonia soda- 
ash, since this is obtained directly as a solid and no evaporation is 

Caustic soda of better quality can be made by Loewig's method, 
but it cannot be made so cheaply as by the use of lime with the tank 
liquor of the Leblanc process, especially in small works where the out- 
put is irregular and uncertain. For although there is no expense for 
lime, and less fuel is used for evaporation in the former method, yet 
an extensive and somewhat costly plant, designed to reduce labor to 
the minimum, is required, and considerable fuel is needed for the 

For the preparation of caustic soda by electrolysis of brine, see 
p. 124. 


In the Leblanc process nearly all the sulphur of the salt-cake re- 
mains in the " tank waste " or residue from the lixiviation of the black- 
ash. The average composition of this waste is shown in the follow- 
ing tables : 







29 20 

29 96 


Na 2 CO 3 .... 

3 16 

1 97 


CaCO 3 




Ca(OH) 2 . . . 








CaS 2 O 3 

1 07 



CaSO 3 


CaSO 4 



CaSiO 3 . . . . 

3 53 

1 34 


Coal - .. 




A1 2 O 3 . 

1 02 




1 65 



Sand /, 




* Chance, J. Soc. Chem. Ind., 1882, p. 266. 








53 14 


CaSO 4 


11 11 

CaSO 3 


3 10 

CaS 2 O 3 




O C\A 

Insoluble in HC1 .... 

10 10 


A1 2 O 3 , Fe 2 O 3 , etc 


11 04 

Water . 



When fresh waste is thrown on the dump, the changes produced 
by weathering cause great nuisance. The air is contaminated by the 
hydrogen sulphide and sulphur dioxide liberated, and the soluble 
polysulphides of calcium and sodium formed are dissolved by rain- 
water, making the objectionable " yellow liquors," which run into 
streams and sewers. 

In fresh waste the sulphur is chiefly in the form of sulphide and 
thiosulphate of calcium, but in weathered material these have been 
converted by oxidation into sulphate and sulphite, which in them- 
selves cause no trouble except by their bulk. 

The simplest method of disposing of waste is to send it out to sea 
and dump it, if the works are so situated that this is convenient ; or, 
if this is impossible, to spread it evenly and beat it down hard to 
prevent as far as possible the infiltration of rain. But since the sul- 
phur thus lost every year represents an enormous money value, many 
attempts have been made to recover it in an available form. Of the 
numerous processes proposed only three need be considered here. 

In Mond's process the waste was treated directly in the lixiviating tanks by 
blowing air or chimney gases through the wet mass. This oxidized the waste 
according to the following reactions : 

1) CaS + 2 H 2 O = Ca(SH) 2 + Ca(OH) 2 . 

2) Ca(SH) 2 + 4 O = CaS 2 O 3 + H 2 O. 

But the hydration and oxidation processes were slow, and after a time it was 
necessary to lixiviate the mass, blow in air, and again lixiviate. By several lixivia- 
tions the calcium sulphydrate and thiosulphate were dissolved, forming " yellow 

* Lunge, Sulphuric Acid and Alkali, 2d ed., Vol. II, p. 815. 


liquors." To recover the sulphur these were treated while still hot with dilute 
hydrochloric acid, the following reactions * taking place : 

3) CaS 2 O 3 + 2 HC1 = CaCl 2 + H 2 O + SO 2 + S. 

4) Ca(SH) 2 + 2 HC1 = CaCl 2 + 2 H 2 S. 

In the presence of the calcium chloride solution the two gases, sulphur dioxide 
and hydrogen sulphide, react upon each other, forming water and free sulphur : 

5) 2 H 2 S + SO 2 = 2 H 2 O + 3 S. 

The hydration and oxidation process was so controlled that the proportion of 
thiosulphate to sulphydrate yielded one molecule of sulphur dioxide to two mole- 
cules of hydrogen sulphide. When properly worked very little escape of hydrogen 
sulphide occurred. The precipitated sulphur was filtered from the solution of cal- 
cium chloride which went to waste. The sulphur was then refined. 

This process recovered about 60 per cent of the total sulphur, but it consumed 
a great deal of hydrochloric acid, which now has considerable value, and some 
sulphur was lost, owing to the formation of sulphate and sulphite of calcium, which, 
being insoluble, were left in the residue after lixiviation. The process is not now in 

Schaffner and Helbig's f process depends upon the reaction between magnesium 
chloride and calcium sulphide in a boiling solution : 

1) CaS + MgCl 2 + H 2 O = CaCl 2 + MgO + H 2 S. 

2) MgO + CaCl 2 + CO 2 = CaCOs + MgCh. 

The second reaction was employed to recover the magnesium chloride, but the 
calcium carbonate formed was too impure for use in the black-ash furnace. The 
hydrogen sulphide set free was pure, and could be utilized by burning it with air, 
and conveying the resulting sulphur dioxide into the lead chambers of the sulphuric 
acid plant ; or the sulphide could be decomposed with sulphur dioxide, according 
to the method given above, reaction (5). Lime-kiln gases were used for the carbon 
dioxide in reaction (2) . This process was not a commercial success. 

The Chance-Glaus process { appears to be the only successful 
method of recovering sulphur on a large scale, and even this has not 
fully realized the original expectations of its promoters. The reac- 
tions of the process were proposed by Gossage in 1837, but although 
he worked on the idea for thirty years, and spent a large fortune in 
experimenting, he failed to make it a success. 

The following are the reactions involved : 

1) 2 CaS + H 2 + CO 2 = CaCO 3 + Ca(SH) 2 . 

2) Ca(SH) 2 + H 2 + CO 2 = CaCO 3 + 2 H 2 S. 

3) CaS + H 2 S = Ca(SH) 2 . 

A carbon dioxide containing at least 30 per cent CO 2 is necessary 
and is obtained in a special form of lime kiln. The tank waste is 

* Mactear proposed to use the same reactions for the treatment of the drain- 
age from old waste heaps, which were creating a nuisance. 

t J. Soc. Chem. Ind., 1882, 264. J Ibid., 1888, 162. 



diluted with water and treated by counter-current system, with the 
carbon dioxide gas, in a series of seven cast-iron cylinders, so arranged 
that one may be emptied and recharged while the others are in un- 
interrupted operation. Since hydrogen sulphide and carbon dioxide 
are acids nearly equal in strength, no hydrogen sulphide is set free 
till the calcium sulphide is converted to a mixture of calcium sul- 
phydrate and bicarbonate. The gas entering a freshly filled tank is 
largely hydrogen sulphide and nitrogen ; the former is absorbed as 
shown in reaction (3), while nitrogen escapes, until the calcium sul- 
phide is nearly all converted to sulphydrate. Reaction (2) then 
progresses and the content of hydrogen sulphide in the gas leaving 
the tank rises rapidly and it is collected in a gasometer. As the treat- 
ment proceeds, the hydrogen sulphide content of the gas falls off while 
the percentage of CO2 rises correspondingly. When the hydrogen 
sulphide is below 30 per cent, the gases are diverted to a freshly filled 
tank, where reaction (3) takes place. 

The hydrogen sulphide collected in the gasometer, together with 
air, is passed through the Claus sulphur kiln (Fig. 49), in which the 
reaction H 2 S + O = H 2 O + S 

takes place. On the grate (A) is a layer of broken fire-brick covered 
with about 12 inches of ferric oxide to serve as a catalyzer of the re- 
action. The mixture of hydrogen sulphide and air is led into the kiln 

at (B), and made 
to pass through the 
ferric oxide (pre- 
viously heated to 
a dull red) ; this 
causes the reaction 
to take place, and 
at the same time 
the heat generated 
by the reaction is 
sufficient to keep 
the iron oxide at the proper temperature, after being once well started. 
Sulphur, nitrogen, and water vapor escape from the kiln. The sulphur 
vapor condenses in the chamber (D) as liquid sulphur, and in (E) as 
flowers of sulphur, while the steam and nitrogen, together with a 
small quantity of sulphur dioxide, pass on to a condensing tower, 
where they are brought into contact with limestone over which water 
is dripping, to retain the sulphur dioxide. When working well, this 


FIG. 49. 


process recovers about 85 per cent of the sulphur. According to 
Lunge, the form of the kiln has been recently modified, but the prin- 
ciple of the process is unchanged. The water in the storage gasom- 
eter is usually covered with a layer of petroleum oil, to prevent the 
absorption of the hydrogen sulphide by the water. 

The process is not very lucrative when the price of sulphur is low, 
but since it reduces the nuisance created by the alkali waste, and 
yields very pure sulphur, a number of English firms employ it. In 
1893, over 30 plants were in operation in England, and more than 
35,000 tons of sulphur recovered. 

For the Parnell and Simpson * process for utilizing alkali waste, 
seep. 111. 


The reactions involved in the ammonia soda process were dis- 
covered by H. G. Dyar and J. Hemming, about 1838, but owing to the 
mechanical difficulties, its practical success was not thoroughly estab- 
lished until 1873. In 1863, Ernest Solvay, a Belgian, constructed an 
apparatus which has led to an enormous development of the industry, 
by which one-half of the world's supply of soda is now made. Its 
advantages lie in the strength and purity of its products and the ab- 
sence of troublesome by-products, such as " tank waste." But it 
does not yield chlorine nor hydrochloric acid, all the former going to 
waste as calcium chloride. 

The ammonia soda process depends upon the fact that sodium 
bicarbonate is but slightly soluble in a cold ammoniacal solution of 
common salt. The technical success of the process depends chiefly 
on the proper regulation of the temperature during the precipitation, 
and on the capacity of the works to handle large quantities of gases 
and liquids. As far as possible, manual labor must be avoided, and 
the products moved and treated in solution or in suspension. The 
reactions are as follows : 

1) NaCl + NH 3 + H 2 + COz = NH 4 C1 + NaHCO 3 . 

2) 2 NH4C1 + Ca(OH) 2 = CaCl 2 + 2 H 2 O + 2 NH 3 . 

The first equation is the chief one ; the second represents the recovery 
of the ammonia, and is essential to the commercial success of the 

The salt is used as a very concentrated brine, which has been 

* J. Soc. Chem. Ind., 1889, 11. 



purified from iron, silica, magnesia, etc. ; it is then saturated with 
ammonia gas, obtained from gas liquors, or by the recovery process 
according to equation (2). The carbon dioxide is obtained partly 
from lime kilns and partly from the calcination of the bicarbonate 
to form the normal carbonate (p. 110). It must contain at least 30 
per cent of CC>2, and is prepared in special forms of continuous lime 
kilns. The lime resulting is used in the recovery of the ammonia 
(reaction 2), and for making caustic soda ; the lime-kiln gases are 
cooled, and the sulphur dioxide removed, by washing in water before 
they pass into the carbonating towers (see below). 

The brine is contained in a tank, under the perforated bottom of 
which the ammonia gas is introduced, and rising through the liquor, 

is rapidly absorbed. The 
heat evolved by the absorp- 
tion is taken up by cold 
water circulating in coils. 
When saturated, the am- 
moniacal brine is pumped 
into a receiving and settling 
tank, from which it is de- 
livered to the " carbonating 
tower" (Fig. 50).* This is 
from 50 to 65 feet high, built 
of cast-iron rings or seg- 
ments (A, A), each about 
3.5 feet high and 6 feet in 
diameter. At the bottom of 
each segment is a flat plate 
having a large hole in the 
centre. Above each plate 
is a dome-shaped diaphragm 
(D) perforated with a great 
number of small holes. In 
modern works a system of 
pipes passes through each segment, as shown at (B, B) ; in these, cold 
water is kept flowing, thus counteracting the heat generated by the 
chemical action. The ammoniacal brine is forced under pressure 
through the pipe (P), entering a little above the middle of the tower, 
which is nearly filled with brine. By this arrangement, any free 
ammonia in the brine, which would be swept away by the stream of 

FIG. 50. 

* After Lunge. 


gases passing up through the tower, is taken up by the carbon dioxide 
in the upper part of the tower. The carbon dioxide, having been pre- 
viously well cooled, is forced through the pipe (C), entering under the 
lowest dome, and rising in small bubbles through the perforations in 
each dome, comes into intimate contact with the ammoniacal brine. 
The bicarbonate of sodium thus precipitated gradually works its way 
down through the tower. A thick, milky liquid, containing the bicar- 
bonate in suspension, and ammonium chloride and common salt in 
solution, is drawn off through (H) at the bottom. 

After a tower has been in use for some days, the holes in the 
domes become clogged with a deposit of bicarbonate crystals, which 
prevent the free passage of the gases. Consequently, every ten days 
or two weeks the liquid must be drawn out and the crystals dissolved 
by filling the tower with hot water or steam. The tower must be 
cooled before starting the process anew. As a rule, several towers 
are employed, so that one may be cleaned and cooled without inter- 
rupting the operation. 

The gases escaping from the top of the tower, consisting princi- 
pally of nitrogen, carbon dioxide, and some ammonia, are passed 
through scrubbers (p. 319), one of which contains brine, which after- 
wards goes to the ammonia saturating tank; in the other is dilute 
sulphuric acid, to absorb the small amount of ammonia which would 
otherwise be lost. The carbon dioxide and nitrogen are allowed to 
escape. The towers are run with the view to the utilization of all 
the ammonia possible, even though there is considerable loss of salt 
and carbon dioxide; usually about one-fourth of the salt remains 

It is now customary to place a smaller carbonating tower in con- 
nection with the large one; in the former the brine is first treated 
with carbon dioxide and the ammonia converted to neutral carbonate 
(NH^COs; then the brine is pumped into the large carbonating 
tower, where it meets more carbon dioxide, and the bicarbonate is 
formed, causing the precipitation of the sodium bicarbonate. More 
heat is liberated in the formation of the neutral carbonate of am- 
monia than in its conversion to the bicarbonate, hence the tempera- 
ture of the precipitation is more easily controlled when two towers 
are used, and less free ammonia escapes with the waste gases. 

A temperature of about 35 C. is most favorable to the formation 
of a granular or crystalline precipitate of bicarbonate, and also to 
the most complete utilization of the ammonia. At higher tempera- 
tures, too much bicarbonate remains dissolved in the liquor; at 


lower temperatures, there is a tendency to the crystallization of am- 
monium acid carbonate and ammonium chloride, while the bicar- 
bonate separates as a very fine precipitate, which is difficult to filter 
from the liquor. 

The milky liquor from the bottom of the tower, containing the 
sodium bicarbonate in suspension, is filtered on sand filters (p. 19) 
connected with a vacuum pump ; or better, it is run into centrifugal 
machines (p. 18), which afford more rapid and complete separation 
of the mother-liquor. The bicarbonate is then washed with water, 
to remove as much of the sodium and ammonium chlorides as possible. 
The mother-liquors and wash waters go to the ammonia recovery 

The sodium bicarbonate is then calcined in large covered cast-iron 
pans or ovens ; this converts the acid salt into soda-ash, and drives 
out any ammonia or moisture still in the mass. The following is 
the reaction : 

2 NaHCQs = Na 2 CO 3 + C0 2 + H 2 O. 

The fumes are passed through coolers and scrubbers to remove 
ammonia; the concentrated carbon dioxide remaining is pumped 
into the carbonating towers. The ammonia liquors go to the ammonia 

A modification of the Thelen pan (Fig. 48, p. 100) is sometimes 
used for this calcining. A gas-tight cover is placed over the pan, 
and the scrapers pass back and forth over the pan bottom, being 
moved by a connecting rod and crank. The gases and steam pass 
off through a pipe set in the cover. In practice, it has been found 
best to leave the mass in this pan only until all the ammonia and 
about 75 per cent of the carbon dioxide of the bicarbonate have been 
expelled ; the calcination is completed in a reverberatory furnace. 

The product of the calcination is called soda-ash ; it is often very 
pure, containing only a trace of salt and a little bicarbonate, and is 
free from caustic soda, sulphide, and sulphate. But its density is 
only 0.8, while that of the Leblanc product is 1.2. This is disad- 
vantageous, owing to the larger packages needed for a given weight 
and to the mechanical loss incurred in operations where the soda-ash 
is exposed to a strong draught of air. In order to increase the den- 
sity, it is sometimes subjected to a second heating in a reverberatory 
(revolving) furnace. 

The second reaction, on p. 107, is that on which the recovery of 
the ammonia depends. The liquid in which the bicarbonate of soda 


was suspended contains undecomposed salt, ammonium chloride, and 
ammonium carbonate. It is passed through an ammonia still, usually 
a tall column or dephlegmator (p. 11). Steam is admitted at the 
bottom of the apparatus, and bubbling up through the liquid, de- 
composes the ammonium carbonate into ammonia, carbon dioxide, 
and water ; the ammonium chloride passes down into the lower part 
of the tower, or the still proper, where it is decomposed by " milk of 
lime." The ammonia set free is cooled and used to saturate the 
brine. The calcium chloride formed remains in solution, and together 
with the excess of salt, goes to waste. Some calcium chloride is re- 
covered and finds use as a dust-layer and binder, on macadam and 
dirt roads. (For various proposals to utilize the waste calcium chloride 
for the production of hydrochloric acid and chlorine, see p. 122.) 

The damp bicarbonate is dried in an atmosphere of carbon dioxide, 
at a temperature of about 90 C. ; this prevents decomposition of the 
sodium bicarbonate, while the ammonium bicarbonate is decomposed, 
the vapors passing to the scrubbers, where the ammonia is recovered. 
A considerable quantity of the bicarbonate of soda is sold directly 
to the manufacturers of baking Dowder and the poorer grades to the 
soda-water makers. 

Caustic soda can be made stronger and purer from ammonia soda- 
ash than from Leblanc ash, and the process is not essentially differ- 
ent, except that no treatment to remove sulphur is necessary ; but it 
cannot be made so cheaply as from the " red liquors " or the " tank 
liquors " of the Leblanc process. If pure lime is used for causticiz- 
ing ammonia soda-a?h, the product is better than in the case of the 
Leblanc ash, as it is free from sulphur, alumina, etc. 

Loewig's process (p. 102) appears especially suited for causticiz- 
ing ammonia soda-ash, since it requires an ash free from silica. 

H. A. Frasch devised a method for caustic soda in which nickel 
hydroxide acts upon sodium chloride in the presence of an excess of am- 
monia. A double nickel-ammonium chloride [Ni(NH 3 )2Cl2 + 4 NHs] 
separates as a violet crystalline mass, leaving caustic soda in the 
solution. This double salt is hygroscopic, dissolves with a blue color, 
and evolves some ammonia. The nickel and ammonia may be re- 
covered by treating the salt with milk of lime. 

The Parnell and Simpson process * was expected to solve the 
problem of the Leblanc " alkali waste," but it has not fulfilled the 
hopes of its promoters. It was proposed to work the two leading 
soda processes in combination. The alkali waste of the Leblanc 

* J. Soc. Chem. Ind., 1889, 11. 


process is boiled in the ammonium chloride liquor from the ammonia 
process, and the vapors of ammonia gas and of ammonium sulphide 
liberated are led directly into a brine solution in the saturating tank. 
The calcium chloride liquor goes to waste. The ammoniacal solution 
of brine and ammonium sulphide produced is sent to a carbonating 
tower, similar to that described on p. 108, and treated with carbon 
dioxide, as in the ammonia-soda process. Sodium bicarbonate is pre- 
cipitated and hydrogen sulphide set free, which may be treated in a 
Claus kiln (p. 106), or burned to sulphur dioxide to use for sulphuric 
acid. The reactions involved are as follows : 

1) CaS + 2 NEUCl = (NH 4 ) 2 S + CaCl 2 .* 

2) (NH 4 ) 2 S + CO 2 + H 2 O = NH 4 HCO 3 + NHJIS. 

3) NH 4 HS + C0 2 + H 2 = NH 4 HCO 3 + H 2 S. 

4) NH 4 HC0 3 + NaCl = NaHCO 3 + NH 4 C1. 

The last three reactions take place f simultaneously in the carbonat- 
ing tower; the hydrogen sulphide generated goes to the sulphur re- 
covery and the ammonium chloride solution carrying sodium bicar- 
bonate in suspension is drawn out and filtered. The ammonium 
chloride liquor is returned to the process. 

The conversion of salt into sodium carbonate by any method in- 
volves an endothermic reaction in some part of the process. Thus 
energy must be expended, necessitating the use of fuel. In the Leblanc 
process, the fuel expenditure is large in carrying out the reactions in 
the salt-cake and the black-ash furnaces. But much of the expended 
energy reappears in the hydrochloric acid, the principal by-product. 

In the ammonia process the principal reactions are exothermic, 
but some fuel is consumed by the calcination of the precipitated bicar- 
bonate and in the preparation of the quicklime used in the ammonia 

* Equation (1) does not exactly represent the facts, as some polysulphides are 
present in the tank waste. 

t According to Lunge, Sulphuric Acid and Alkali, 3d ed., Vol. Ill, p. 204, the 
sodium bicarbonate is formed by agitating the brine with crystallized ammonium 
bicarbonate, the latter being obtained by saturating the ammonium sulphide solu- 
tion with carbon dioxide. The carbon dioxide, which must be very pure and con- 
centrated, is made by heating ammonium bicarbonate crystals to 74 C., in a retort, 
CO2, steam, and NHs passing off. By scrubbing (p. 319), the carbon dioxide is ob- 
tained pure. 

Ammonium bicarbonate is also prepared by passing lime-kiln gases into a solu- 
tion of ammonia or neutral ammonium carbonate, and then cooling it to crystallize 
the bicarbonate. 


recovery and for generating carbon dioxide. Although less fuel is 
used than in the Leblanc process, the practical economy of the am- 
monia process is not so great as would at first appear ; for all the 
chlorine is lost, together with much of the original salt used. As a 
method for soda-ash it is far superior to the Leblanc, but until a prac- 
tical process for the cheap production of chlorine is discovered, the 
latter will continue to be an extensive industry. 


Cryolite is a double fluoride of sodium and aluminum, found as 
a mineral in southern Greenland. As no other important deposit 
has been found, the supply is limited, and only two or three manu- 
factories using this process were established, one of them in this 
country. The reactions involved are as follows : 

1) A1F 3 - 3 NaF + 3 CaCO 3 = NaAlO 2 + Na 2 O + 3 CaF 2 + 3 CO 2 . 

2) NaAlO 2 + Na 2 O = Na 3 AlO 3 . 

3) 2 Na 3 A10 3 + 3 H 2 O + 3 CO 2 = 3 Na 2 CO 3 + 2 A1(OH) 3 . 

The ground cryolite is mixed with powdered limestone, and calcined 
at a red heat. Carbon dioxide escapes, and a mixture of calcium 
fluoride, sodium oxide, and sodium aluminate remains. On lixiviat- 
ing this mixture with water, another sodium aluminate is formed and 
goes into solution, leaving the calcium fluoride as an insoluble resi- 
due. The solution of sodium aluminate is then decomposed according 
to the third reaction, by passing into it purified lime-kiln gases, or 
the furnace gases of the calcining operation. Hydrated alumina is 
precipitated, while sodium carbonate remains in solution. Sal-soda 
may be made by evaporating the solution, and was formerly the chief 
source of bicarbonate for culinary and medicinal purposes. If carried 
to complete dryness and calcined, a high grade of soda-ash is obtained. 
By causticizing, it yields a very excellent caustic. 

The by-products aluminum hydroxide and calcium fluoride are 
used in the alum and glass industries respectively. 

Many other processes for the manufacture of soda from salt have 
been proposed, but none of them are now of any commercial impor- 
tance. A small amount of soda is still made from kelp or varec, 
which is the ash of seaweeds. 

A new process for making soda has been proposed,* which is in- 
teresting, but has not as yet been placed on a practical basis. Salt- 

* J. Soc. Chem. Ind., 1895, 933. 


cake is made from salt by the Hargreaves process (p. 92) ; then in 
the same cylinder and at the same temperature, it is treated with 
water gas. This reduces the salt-cake to sodium sulphide, while water, 
carbon monoxide, and hydrogen escape. These vapors are cooled, the 
water condensed, and the mixture of gases burned, the products of 
combustion, carbon dioxide and water, passing into the cylinders con- 
taining the sodium sulphide. Hydrogen sulphide and sodium car- 
bonate are formed, and as the temperature is much above 100 C., no 
water can combine with the carbonate. The hydrogen sulphide is 
burned to sulphur dioxide, and the latter returned to the Hargreaves 
process. The reactions involved are as follows : 

1) 2 NaCl + SO 2 + H 2 O + O = Na 2 SO^+ 2 HC1. 

2) NajSOfc.+ 5 CO + 5 H 2 = Na 2 S + 4 H 2 O + 5 CO + H 2 . 

3) CO + H 2 + O 2 = CO 2 + H 2 O. 

4) Na 2 S + C0 2 + H 2 = Na 2 CO 3 + H 2 S. 

5) H 2 S + 3 O = H 2 O + SO 2 . 

This process seems to offer several advantages, viz. : 

1. Cheap materials. 

2. Small outlay for labor, the materials not being handled 
from the time the salt is charged into the cylinders until the soda- 
ash is raked out. 

3. No waste products nor nuisance. 

4. The temperature constantly decreases, being highest when the 
furnace is charged and lowest when the soda-ash is finished. 

5. The process yields hydrochloric acid which can be utilized. 
For the methods of producing caustic soda and chlorine by elec- 
trolysis of brine, see Chlorine, p. 124. 


Berichte ueber die Entwickelung der chemischen Industrie. A. W. Hoff- 
mann, Vol. I, 418. (1875.) 

History, Products, and Processes of the Alkali Trade. Charles T. King- 
zett, London, 1877. (Longmans.) 

Manual of Alkali Trade. John Lomas, London. (Crosby, Lockwood Co.) 

Die Fabrikation der Soda nach dem Ammoniak-Verfahren. H. Schreib, 
Berlin, 1905. 

Sulphuric Acid and Alkali. G. Lunge, 3d ed., Vols. II, 1909, III, 1911. 
(D. Van Nostrand Co., New York.) 

J. Soc. Chem. Ind : 

1883, 405, Walter Weldon. 1885, 527, Ludwig Mond. 1886, 412, 

E. K. Muspratt. 

1887, 416, Watson Smith. 1888, 162, Alexander Chance. 1889, 11, 
E. Parnell. 


Chlorine is extensively used in the arts as a bleaching and oxidiz- 
ing agent. It is chiefly employed in the form of a solution of " bleach- 
ing powder " or " chloride of lime," which contains calcium hypo- 
chlorite, and as chlorates or hypochlorites of the alkali metals. Liquid 
chlorine, compressed in steel cylinders, has recently become an article 
of commerce, and this form of shipment may be extended in the future. 

Practically all the chlorine used in the arts must be derived from 
the chlorides of sodium, potassium, or magnesium, which are found 
more or less abundantly in nature. A large part of the hydrochloric 
acid made from salt (p. 88) is used for making chlorine. Since this 
acid is the chief by-product of the Leblanc process, a plant for mak- 
ing bleaching powder is always a part of those works. 

The important methods of making chlorine from the acid may 
be considered under two heads : ':hose using manganese oxides for 
decomposing the acid, and those not using manganese for this purpose. 

The function of manganese is to oxidize the hydrogen of the acid, 
forming water and liberating the chlorine. At the same time, the 
manganese is converted into chloride, and being expensive, its recovery 
in a form that permits of its return to the process is essential. 

The oxides of manganese are found in nature as pyrolusite (MnO 2 ), 
braunite (Mn 2 O 3 ), manganite (Mn 2 O 3 H 2 O), hausmannite (Mn 3 O 4 ), 
wad, and psilomelane, the last two of indefinite composition. The 
reactions occurring when manganese oxides are treated with hydro- 
chloric acid are as follows : 

1) MnO + 2 HC1 = MnCl 2 + H 2 O. 

2) MnO 2 + 4 HC1 = MnCl 2 + 2 H 2 O + 2 Cl. 

3) Mn 2 3 + 6 HC1 = 2 MnCl 2 + 3 H 2 O + 2 CL 

4) Mn 3 O 4 + 8 HC1 = 3 MnCl 2 + 4 H 2 O + 2 Cl. 

Thus it is readily seen that with pyrolusite, less acid is necessary 
for a given yield of chlorine, and a smaller quantity of manganous 
chloride must be treated to recover the manganese. This ore is 
purchased according to its, content of MnO 2 , which is estimated by 
determining the " available " oxygen. The presence of iron oxides, 
silica, calcium carbonate, etc., is disadvantageous. 

In small works, especially where no attempt is made to recover 
the manganese, the process is carried on in simple stills of earthen- 




FIG. 51. 

ware or sandstone. The earthenware stills 
(Fig. 51) * are cheap, but of limited capacity. 
They are heated by blowing free steam into 
the wooden casing in which they are set. 
The pyrolusite is put into the central per- 
forated cylinder, and the acid runs through 
the pipe (A), chlorine escaping at (B). 
Sandstone stills (Fig. 52) * are made from 
single blocks of sandstone, or built up of 
slabs, the joints being made tight by a 
rubber packing, or by a lute of clay and 
linseed oil. The pyrolusite rests on a false 
bottom (A), and the acid is run in through (B), while steam is blown 
in through the sandstone pipe (C). Chlorine escapes through (D). 
These stills are 
larger than the 
earthenware ones, 
but do not utilize 
the acid so com- 

The chlorine is 
conducted through 
pipes of lead or 
earthenware, or 
tubes coated with 
bakelite enamel. Since valves in these pipes are rapidly corroded, 
a device shown in Fig. 53 * is used to shut off the flow of gas. A 

U-shaped bend is made in the pipe, and 
a small flexible tube attached at the lowest 
point of the U, connecting it with the 
vessel (A), filled with water. By raising 
(A), the water flows into and fills the 
U-pipe to the line (CD), cutting off the 
flow of gas. By lowering (A) to (A')> 
the water runs out of the U, and the flow 
of gas is uninterrupted. 

The liquor remaining in the still con- 
tains much free acid, manganous chloride, 
ferric chloride, etc. It continues to evolve 
some chlorine for a long time, and is a 

FIG. 52. 

* After Lunge. 



very offensive and troublesome material to dispose of, since it pol- 
lutes the air, or the streams, into which it passes. 

Of the many attempts to recover the manganese, the two follow- 
ing are the most important : 

By Dunlop's method, the " still liquor " is neutralized cold, with 
powdered limestone, until all free acid is removed and the iron pre- 
cipitated. The clear solution of manganous and calcium chlorides is 
then mixed with a carefully determined quantity of powdered lime- 
stone or chalk, and heated under pressure by steam. This precipi- 
tates the manganese as carbonate, which is settled, and the solution 
of calcium chloride drawn off. The manganous carbonate is washed 
and then calcined at about 300 C. in a retort, while water spray and 
a current of air are introduced. This produces a mixture of MnC>2, 
MnO, Mn 2 O 3 , etc., containing 
about 70 per cent of the diox- 
ide. The process requires an 
expensive plant and consumes 
much fuel. 

The Weldon process * for 
manganese recovery is the 
most successful, and is in use 
in many large works, since it 
furnishes a continuous pro- 
cess for chlorine making and 
manganese recovery. The 
"still liquors" are neutral- 
ized with just sufficient pow- 
dered limestone or chalk to 
remove free acid and precipi- 
tate the iron. This is done 
in the tank (A) (Fig. 54), f 
provided with a stirrer. The 
mixture is then pumped into settling tanks (B, B), where the precipi- 
tate deposits. The clear solution of manganous and calcium chlorides 
is then drawn into the " oxidizers " (C), where steam is blown in to 
heat it to 55 C. Milk of lime made from pure lime, especially free 
from magnesia, is added from (E) until tests show that the manganese 
is all precipitated; meanwhile air is slowly forced into (C). The 
quantity of " milk " used is noted, and then from one-half to one- 

FIG. 54. 

* J. Soc. Chem. Ind., 1885, 525. 

t After Lunge. 


quarter more is added, and the air blast turned on at full strength. 
This addition of an excess of lime is necessary to hasten and com- 
plete the conversion of manganous hydroxide into the peroxide, and 
to prevent the formation of Mn 3 O 4 (" red batch "). The total quan- 
tity of lime used should be such that the precipitate formed during 
the blowing contains approximately two molecules of manganese per- 
oxide to one of calcium oxide. This is the so-called " acid calcium 
manganite " (CaO MnCy -f- (MnO MnC^), a mixture of mangan- 
ites of calcium and manganese. It forms a thin, slimy, black mass, 
and is called "Weldon mud." By adding a little more neutralized 
" still liquors " during the " blowing," some of the calcium oxide in 
the calcium manganite can be replaced by manganese from the man- 
ganous chloride of these liquors. 

The calcium chloride liquor, in which the mud is suspended, is 
run into settling tanks (D, D), from which the supernatant solution 
is drawn off as waste. The Weldon mud is then run into the chlorine 
stills (F, F) as a thin paste ; if of good quality, it contains about 80 
per cent of its manganese as MnC>2, and owing to its fine state of 
division, is readily decomposed by dilute hydrochloric acid. 

A small loss of manganese occurs in the precipitate from the first 
neutralization with marble or chalk dust; this loss is made up by 
decomposing some pyrolusite with hydrochloric acid in a small still 
(G), and adding this liquor to that from the stills (F, F). 

The Weldon process works continuously and almost automat- 
ically, the materials being handled by pumps as liquids or slimes. It 
is also very rapid, producing large amounts of chlorine, with but 
slight loss (2 to 3 per cent) of manganese oxide. But even at its 
best, only about one-third of the chlorine of the hydrochloric acid is 
obtained as gas, the remainder going to waste as calcium chloride in 
the liquor from the oxidizers. 

Deacon's process * is the most successful chemical method of 
producing chlorine without the use of manganese. It depends on 
the oxidation of hydrochloric acid gas by the oxygen of the air. This 
is done in the presence of certain metallic salts, which may act as 
" contact " substances, or as carriers of oxygen from the air to the 
acid, the apparent reaction being : 

4 HC1 + 2 = 2 H 2 + 2 C1 2 . 

The most satisfactory " contact " or " catalytic " substance for this 
purpose is copper chloride. When cupric chloride is heated to 400 

* Chemical News 22 (1870), .157. 



C., it dissociates into cuprous chloride and free chlorine. Then, on 
exposing the cuprous chloride to oxygen, cupric oxide is formed and 
more chlorine set free. But the cupric oxide, reacting with hydro- 
chloric acid gas, forms water and cupric chloride. The following are 
the reactions involved : 

1) 2 CuCl 2 = Cu 2 Cl 2 + C1 2 . 

2) Cu 2 Cl 2 + O 2 = 9 CuO + C1 2 . 

3) 2 CuO + 4 HC1 = 2 CuCl 2 + 2 H 2 O. 

Thus the catalytic substance is regenerated and the cycle of changes 
begins anew. 

During the dissociation of cupric chloride 32 Calories is absorbed, 
but in the other reactions 60.4 Calories is evolved. Hence there is 
a gain of 28.4 Calories, and theoretically the process once under way 
no addition of heat is needed. But, in fact, owing to losses by radia- 
tion, convection, and conduction, some heat must be supplied, and 
the mixture of air and hydrochloric acid gas is heated to 400 C. be- 
fore admitting it to the " decomposers." Since the reaction between 
the hydrochloric acid and the oxygen is reversible, an equilib- 
rium is established, and so all of the chlorine is not recovered. 

The plant for the process (Fig. 55) * is quite extensive. The 
gases from the salt-cake pan (A),f together with air, are passed 
through cooling pipes and drying tower (B) to condense 
moisture ; then they go through the " superheater " (C), 
where the temperature is raised to 400 C. The hot gases 
then pass into the "decomposer" (D), a tall cast-iron 

FIG. 55. 

cylinder, containing bits of brick or other porous material which have 
been soaked in a solution of cupric chloride. Here the above reac- 
tions take place, and the resulting mixture of chlorine, hydrochloric 
acid, nitrogen, steam, and oxygen, passes through a condensing appa- 

* After Lunge. 

t Roaster gas is too dilute and impure. 


ratus (E, E) to remove the hydrochloric acid, and then through a 
coke tower (F, F) sprinkled with concentrated sulphuric acid to 
remove all the moisture ; finally, the dry chlorine gas (with the 
nitrogen and oxygen) goes to the chambers where bleaching powder 
is made (p. 132). 

The catalytic substance in the decomposer becomes inactive after 
a time (it seldom lasts more than four months) and must be replaced 
by fresh material. To accomplish this without interrupting the pro- 
cess the decomposers are built in separate compartments, each holding 
about six tons of broken brick; every two weeks one compartment 
is emptied and recharged without discontinuing the flow of gas 
through the others. This loss of activity in the catalytic substance 
is attributed * to the presence of sulphuric acid in the gases from 
the salt-cake furnace. To overcome this difficulty, Hasenclever 
devised a method f by which an aqueous solution of impure hydro- 
chloric acid, made in the bombonnes and coke towers, is run into hot, 
concentrated sulphuric acid (1.42 Tw.) while a blast of air is forced 
through the mixture. The sulphuric acid absorbs the water and 
generates pure HCl gas, which mixes with the air in proper proportion 
for use in the decomposer of Deacon's process. By this method, 
84 per cent of the hydrochloric acid gas is decomposed according to 

the reaction : 

4 HCl + O 2 = 2 H 2 O + 2 C1 2 . 

The diluted sulphuric acid is concentrated and returned to the 
process. The dilute hydrochloric acid which passes through the 
apparatus is recovered by washing the chlorine gas, and is mixed 
with the strong acid from the roasters. 

Owing to the admixture of nitrogen with the chlorine, the latter 
is weaker than that furnished by the Weldon process and for making 
bleaching powder a special form of absorption chamber must be used. 

When the hydrochloric acid gas is taken directly from the salt- 
cake pan or from the muffle furnace, there is apt to be some difficulty 
in working Deacon's process, owing to the variation in the rate of 
liberation of the gas. Much care in the regulation of the air supply 
is necessary. 

The hydrochloric acid gas from the Hargreaves process (p. 92) is 
too dilute for direct use in the Deacon apparatus. 

Arsenic in the sulphuric acid used in the salt-cake pan, or for dry- 
ing the chlorine gas, causes a loss, in the first case by rendering the 

* Berichte d. chem. Gesellschaft, IX, 1070. 
t Lunge, Sulphuric Acid and Alkali, II, 417. 


copper salt inactive, and in the second, by forming hydrochloric acid, 

thus : 

As 2 O 3 + 4 Cl + 2 H 2 O = As 2 O 5 + 4 HC1. 

Part of this hydrochloric acid combines with the As2Os to form 
a solution which condenses in the pipes between the drying tower 
and the bleaching powder chambers. But some of the acid is left in 
the chlorine and attacks the bleaching powder, causing it to be 
" weak." 

The cost of a Deacon plant is rather more than of a Weldon plant 
of the same capacity ; and while it is theoretically a superior process 
and requires less labor, it is not yet in general use. 

Several processes for the preparation of chlorine by the use of 
nitric and sulphuric acids have been proposed. 

Schloesing's process * for chlorine by the use of nitric and hydrochloric acids 
and manganese oxides depends upon the following reactions : 

2 HC1 + 2 HNO 3 + 4 MnO 2 = Mn(NO 3 ) 2 + 2 H 2 O + C1 2 . 

The reaction is carried out by heating the mixture of acids and manganese 
peroxide to 125 C., using an excess of nitric acid. By heating the manganous 
nitrate to 180 to 190 C., it is decomposed, and nitric acid may be regenerated from 
the vapors by treating them with air and steam, while manganese peroxide is re- 

Mn(N0 3 ) 2 = Mn0 2 + N 2 O 4 ; 
H 2 O +O=2 HNOa. 

Wischin, Just, and Alsberge have each patented modifications of the above 
process. Alsberge proposes to apply the method to the recovery of chlorine from 
the ammonium chloride liquors of the ammonia soda process, by employing the 
following equations : 

1) 2 NH 4 C1 + MgO + MnO 2 = MgCl 2 + MnO 2 + H 2 O + 2 NH 3 . 

2) MgCl 2 + MnO 2 + 4 HNO 3 = 'Mg(NO 3 ) 2 + Mn(NO 3 ) 2 + 2 H 2 O + Clt. 

By evaporating to dryness and calcining the residue, the nitrates are decom- 
posed thus : 

Mg(NOs)* + Mn(NO 3 ) 2 = MgO + MnO 2 + 2 N 2 O 4 + O. 

The peroxide of nitrogen is converted to nitric acid by treatment with steam 
ana air : " N 2 4 + H 2 + O = 2 HN0 3 . 

Dunlop's nitric acid-chlorine process depends t upon one or the other of the 
following equations : 

2 NaCl + 2 NaNO 3 + 4 H 2 SO 4 = 4 NaHSO 4 + N 2 O 4 + Ch + 2 H 2 O. 
4 NaCl + 2 NaNO 8 + 6 H*SO 4 = 6 NaHSO 4 + N a O + 2 Cl + 3 H 2 O. 

* Zeit. angew. Chemie, 1893, 99, Lunge and Pret. Wagner's Jahresbericht, 
1862, 235. 

t Lunge, Sulphuric Acid and Alkali, III, 508. 


The mixture of salt, sodium nitrate, and sulphuric acid is heated in an iron 
cylinder which is surrounded by the flames of the fire. The vapors leaving the re- 
tort are passed through concentrated sulphuric acid which retains the nitrogen 
oxides, and the chlorine is then washed with water to remove any traces of hydro- 
chloric acid. 

The nitrous vitriol obtained may be used in the sulphuric acid manufacture. 
The process was worked on a large scale at St. Rollox, England, but has been 

Donald's process * consists in passing the hydrochloric acid vapor from a salt- 
cake furnace through sulphuric acid to dry it, and then through a mixture of nitric 
and sulphuric acids kept at C., when the following reactions take place: 

2 HC1 + 2 HN0 3 = 2 H 2 O + N 2 O 4 + C1 2 . 

. The gas mixture thus formed is led through dilute nitric acid, when the follow- 
ing takes place : - + ^ = HNQ> + 

By passing through concentrated sulphuric acid, the nitrous acid and nitrogen 
oxides are absorbed, while the chlorine is sent to the bleaching powder chambers. 

Many attempts have been made to recover the chlorine from the 
waste liquors of the ammonia soda process, but no one of them has 
yet proved a commercial success. Several of them are, however, 
interesting, and deserve a few words. 

Solvay conducted elaborate experiments in which he tried to 
realize the reaction : 

CaCl 2 + SiO- + O = CaSiO 3 + C1 2 . 

But calcium chloride is very stable, and its decomposition in this 
way is incomplete, and requires enormous expenditure of heat, be- 
sides that used in evaporating the solution of calcium chloride to 

Magnesium chloride is more easily decomposed than calcium 
chloride, and several processes have been devised, based on the use 
of this salt. It is proposed to use magnesium oxide or hydroxide 
instead of lime for decomposing the ammonium chloride solution of 
the ammonia process; by this, magnesium chloride is formed and 
the ammonia gas set free. Both Solvay and Weldon, within a few 
days of each other, patented methods for carrying out this idea. 
But the reaction between ammonium chloride and magnesia is not 
complete, and the solution of magnesium chloride obtained is dilute. 
Viewed as a method for chlorine, more promising results were ob- 
tained by using the concentrated magnesium chloride mother-liquors 
from the Stassfurt industries (p. 161), or from other manufacturing 
operations. The magnesium chloride solution is evaporated to dry- 

* Lunge, Sulphuric Acid and Alkali, III, 514. 


ness at a very low temperature, and the dried chloride is decomppsed 
by passing air or steam over it while heated to a red heat. The re- 
actions are as follows : 

1) MgCl 2 + O = MgO + C1 2 . 

2) MgCl 2 + H 2 O = MgO + 2 HC1. 

The hydrochloric acid obtained is used in the Weldon or Deacon 

The Weldon-Pechiney * process was the most successful of the 
magnesia methods, though none of them can be said to be profitable. 
In this, magnesium chloride solution (made by dissolving the oxide 
in hydrochloric acid, or obtained from waste liquors) is concentrated 
until it contains six molecules of water for each molecule of magnesium 
chloride; then 1 equivalents of magnesium oxide are stirred into the 
solution. The pasty mass heats and soon hardens to a solid cake of 
magnesium oxy chloride, which is broken into lumps about the size of 
a butternut, and screened to remove the dust. The presence of dust 
causes the mass to cake badly during the subsequent drying. The lumps 
are dried at a temperature not exceeding 300 C., by passing a current 
of hot air over them while spread in a thin layer on gratings. Too high 
temperature causes a loss of chlorine as such. If not thoroughly dried, 
chlorine is lost as hydrochloric acid. The dried oxychloride is quickly 
decomposed in a special form of retort, which has been heated by pro- 
ducer gas to a temperature of 1000 C. before the charge is introduced. 
Air is passed into the retort to assist in the decomposition, which must 
be rapid, or the yield of chlorine is reduced. Magnesium oxide is left in 
the retort, while a mixture of chlorine, hydrochloric acid, and nitrogen 
escapes. The hydrochloric acid is recovered by washing the gases with 
water, and is used to dissolve part of the oxide from the retort. The 
chlorine, mixed with nitrogen, is used for bleaching powder, or for other 
purposes. The residue of magnesium oxide is returned to the first stage of 
the process. 

The yield, including the hydrochloric acid recovered, is about 88 per 
cent of the whole amount of chlorine in the magnesium chloride. About 
40 per cent is obtained as free chlorine, and 48.5 per cent is returned to 
the process as MgCl 2 and HC1. 

Of the several methods that have been devised for the direct produc- 
tion of chlorine from the ammonium chloride formed in the ammonia 
soda industry, Mond's process,! which provides for the recovery of the 
ammonia, has been most carefully developed, but its practical success 
is as yet problematical. It is based on the dissociation of ammonium 
chloride into ammonia and hydrochloric acid, at a temperature of 350- 
360 C. ; the hydrochloric aci.d being then combined with some metallic 
oxide, to form a non-volatile chloride, to be later decomposed with libera- 

* .1. Soc. Chem. Ind., 1887, 775. 

t Chemische Industrie, 1892, 466. J. Soc. Chem. Ind., 1887, 140, 216, 217, 440 ; 
1888, 626, 845, 


tion of the chlorine. Oxide of nickel was used at first, but was later 
abandoned in favor of magnesium oxide. The reactions are : 

1) MgO + (2 NH 3 + 2 HC1) = MgCl 2 + H 2 O + 2 NH 3 . 

2) MgCl 2 + O = MgO + C1 2 . 

Since an excess of magnesia is present, it is probable that considerable 
magnesium oxy chloride is also formed, according to the reaction : 

2 MgO + 2 HC1 = MgO MgCl 2 + H 2 O. 

Then this is decomposed by the ah* (reaction 2), thus : 
MgO MgCl 2 + O=2 MgO + C1 2 . 

The liberated ammonia passes from the apparatus to the scrubbers 
of the ammonia recovery process. The complete recovery of this am- 
monia is the first essential to the success of this method. 

The ammonium chloride is crystallized from the liquors of the Solvay 
carbonating towers (p. 108), by cooling them to about C. The dry 
crystals are then vaporized by introducing them into melted zinc chloride, 
contained in an iron vessel lined with an antimony alloy. 

The magnesium oxide, mixed with some potassium chloride, china 
clay, and lime, is made into balls (" pills "), about one-half inch in diame- 
ter, and baked. The decomposer is then filled with the " pills " and 
heated to 360 C., when vapors of ammonium chloride are passed through 
the apparatus. The reaction between the ammonium chloride and 
magnesia raises the temperature in the decomposer above 400 C. Next, 
inert gases, such as those from lime kilns, heated to 550 C., are passed 
into the apparatus to drive out the ammonia and water vapors ; these 
also heat the charge above 500 C. Air, heated to 800 C., is then ad- 
mitted to break up the magnesium chloride (reaction 2) and regenerate 
the oxide; it also sweeps out the chlorine formed. After cooling to 
360 C. ammonium chloride vapors are again introduced and the cycle 
of operations is repeated. To secure uninterrupted working, there are 
usually four decomposers in each plant. 



By passing a current of electricity through a sodium chloride 
solution the salt is decomposed into chlorine at the anode and sodium 
at the cathode. But the latter at once decomposes a molecule of 
water of the solution, forming caustic soda and setting free hydrogen. 
Hence the products of electrolysis are chlorine, caustic soda, and 
hydrogen, of which the last mentioned is of slight value at present. 

While electrolysis appears very simple and direct at first glance, 
there are, in fact, serious difficulties encountered in all electrolytic 
processes for decomposing salt. The migration velocity of hydroxyl 
ions is so much greater than that of chlorine ions, that the former 


carry a large part of the current, which tends to an accumulation of 
the hydroxyl ions in the anode compartment, where various reactions 
take place, resulting in the liberation of some oxygen, which mixes 
with the free chlorine, or attacks the carbon anodes, forming carbon 
dioxide in the gas. This deposition of hydroxyl ions also represents 
a serious waste of energy. The chlorine diffuses somewhat, through 
the electrolyte, and coming in contact with the caustic from the 
cathode, increases the tendency to secondary reactions. 

To prevent this migration and diffusion, various devices have been 
proposed, and the great number of cells devised to overcome these 
difficulties may be brought under four general classes : 

I. Those in which the products of the electrolysis are kept apart 
by use of a porous diaphragm or partition, in the cell. 

II. Those employing a moving mercury cathode to remove the 
alkali metal from the immediate field of decomposition. 

III. Those depending upon the specific gravity of the alkali solu- 
tion produced, to keep it away from the action of the chlorine. 

IV. Those using fused salt as electrolyte, thus avoiding secondary 
reactions by eliminating the hydroxyl ion from the bath. 

Porous diaphragms between the anode and cathode seem simple, 
but no material is available which offers no resistance to the passage 
of electricity, yet prevents the migration and diffusion. Furthermore 
very few substances can be used for the diaphragms, because of the 
destructive action of the chlorine. Then, magnesia, silica, etc., from 
impurities in the salt, deposit in the pores of the diaphragm, and with 
continued working of the cell cause a considerable increase of the 
resistance. The nascent chlorine is also very destructive to the 
anode, and only platinum, or fused magnetite (FesO^ or iron oxide, 
which are expensive and fragile, or Acheson graphite have proved 
efficient in withstanding its action. 

If the hydrogen liberated at the cathode is permitted to escape 
through the solution, it stirs the liquid, aiding the diffusion of the 
chlorine, and the consequent formation of chlorates and hypochlorites, 

1) NaCl = Na+Cl. 

2) Na + H 2 O = NaOH + H. 

3) 2 NaOH + 2 Cl = NaCIO + NaCl + H 2 O. 

4) 3 NaCIO = NaClO 3 + 2 NaCl. 

5) NaCIO + 2 H = NaCl -1- H 2 O. 

Reactions (3), (4), and (5) cause loss, since they regenerate salt. 



Le Sueur's process,* one of the earliest of the diaphragm methods, 
has undergone several modifications. The diaphragm is asbestos 
cloth which rests against the iron gauze cathode. The anodes, 60 
in each cell, are platinum wire gauze. Diffusion of the sodium hy- 
droxide through the diaphragm is hindered by feeding salt solution 
slowly into the anode chamber, thus keeping the level of liquid in 
that space slightly higher than in the cathode compartment. Each 
cell takes 1000 amperes, at 6| volts, and the chlorine efficiency is 
claimed to be 88 per cent or higher ; but the caustic production is 
less efficient. 

In Carmichaers apparatus f the asbestos diaphragm, impregnated 
with Portland cement, rests on the horizontal or slanted cathode at 
the bottom of the cell ; above it is a bell to collect the hydrogen given 
off. The anode, suspended in the top of the cell, is a grating of 
copper rods, covered with hard rubber, through which numerous plati- 
num points project into the brine. The brine is fed in at the top of 
the cell in a rapid stream of drops, so regulated that the caustic 
formed at the cathode is drawn off before it has time to diffuse through 
the liquid. The cathode liquor produced contains about 20 per cent 
caustic soda, and about 75 per cent of the salt is decomposed. The 
reaction is conducted at about 80 C., in the top of the cell, near the 
anode, while the region around the cathode is kept as cool as possible. 
The Hargreaves-Bird cell | Fig. (56) consists 
pf a tall, narrow, vertical cell, having two up- 
right diaphragms (B, B) composed of Portland 
cement with asbestos, supported on the vertical 
cathodes of iron wire gauze. The cell is 8 feet 
long by 6 feet high and about 14 inches thick, 
and is enclosed in a cast-iron box, which also 
supports the cathodes. The anode space (D, D) 
between the diaphragms is filled with strong 
brine, and the weakened brine leaves the com- 
partment through an overflow pipe near the 
top of the cell. By admitting steam to the 
cathode space (C, C), between the outside case 
and the diaphragms, the sodium ions passing through them are com- 
bined to form caustic soda, which washes down to the outlet pipe. 

*J. Soc. Chem. Ind., 1892, 963; 1894, 453. 

J. Am. Chem. Soc., 1898, 868. U. S. Pat. No. 723,398. 
t Zeitschr. f . angew. Chemie, 1896,. 537. 
J J. Soc. Chem. Ind., 1894, 250, 256; 1895, 166, 1011. 

Eng. Min. J., 1898 (65), 611 ; 1902 (73), 471, 

FIG. 56. 



If carbon dioxide or furnace gases are admitted to the cathode space, 
sodium carbonate solution is formed; this improves the efficiency, 
since the rate of migration of the carbonate ion is less than that of 
the hydroxyl ion. The anodes (A) are graphitized carbon blocks, 
connected by copper rods, and buried in cement so that only the 
ends of the carbons are exposed to the brine. Eight anodes are 
placed in each cell. 

The Townsend cell * (Fig. 57) also has vertical diaphragms (D) 
of asbestos cloth, the pores of which are filled with a mixture of iron 
oxide, asbestos fibre, and precipitated amorphous iron hydroxide. 
This is supported on the perforated iron plate cath- 
odes (S, S), and forms a chamber filled with brine 
surrounding the graphitized carbon anodes (G). 
The cathode compartment is filled with kerosene 
oil (K), and the drops of caustic liquor percolat- 
ing through the diaphragm, on meeting the oil, 
assume the spherical shape and fall to the bottom 
of the cathode space, under the oil ; the caustic 
liquor then escapes by a trap (A). The brine (T) 
is continuously added to the anode space. These 
cells take from 2500 to 5000 amperes at an aver- 
age of 4.7 volts. The caustic liquor averages 150 
grams sodium hydroxide and 200 grams of salt, 
per litre. 

The Griesheim-Elektron process | (Fig, 58) employs porous 
cement diaphragms forming the walls of the anode compartment; 

the anodes (A) are 
cast from fused iron 
oxide. From six to 
twelve of these small 
cells are placed in an 
iron vat (K) consti- 
tuting the cathode 
bath, and provided 
FlG - ** with a run-off pipe 

for the caustic liquor. Salt solution is introduced into the anode 
chamber through a pipe (Z). 

* Electro-chem. Met. Ind., 1907, 209. J. Soc. Chem. Ind., 1907, 746. 

7th Internat. Cong. Appd. Ghem. 1909, Sec. X, p. 36. 
fBer. 1909 (42), 2897. 
Elektrochemie wassriger Losungen. F. Forster, Leipzig, 1905. 

FIG. 57. 



The Castner process * was the first successful method employing 
a moving mercury cathode. The cell (Fig. 59) is divided into three 
compartments and is given a slight rocking motion by the cam (E). 

The two outside compartments 

| I ~ contain the iron cathodes, and 

the centre chamber contains the 
carbon anodes. The mercury 
lies about one-eighth inch deep 
on the bottom of the cell, and 
flows alternately from each anode 
compartment into the cathode 

FIG. 59. ... 

chamber, where it is depnved or 

its sodium content by the water therein, forming caustic liquor. 
Brine is fed into the anode compartment, and the sodium set free 
is taken up by the mercury, forming amalgam, which flows into 
the cathode compartment. Owing to a slight reaction between the 
sodium and chlorine in the anode chamber, the amalgam is a little 
deficient in sodium ions in the cathode chamber, and the current 
tends to decompose water, thus oxidizing the mercury. To avoid 
this, a metallic connection is made outside the cell, between the 
iron cathode and the mercury-amalgam layer in the caustic chamber ; 
this accelerates the reaction between the water and the amalgam, 
and yet is independent of the current in the chlorine chamber. 
Kellner has improved on the 
Castner process by making 
the decomposing cell sta- 
tionary and circulating the 
mercury through the anode 
chamber by means of an 
Archimedian screw, or wheel 
pump. The anodes of plati- 
num gauze weigh about 1 
gram each, and 525 are 
placed in each cell. The 
cells take 4000 to 10,000 
amperes at 4.3 volts, thus 
reducing the size of plant as compared with the original cells. 

The Whiting cell f (Fig. 60) operates intermittently, the mercury 
remaining stationary in the anode compartment until an amalgam of 

FIG. 60. 

* J. Soc. Chem. Ind., 1893, 301. Eng. Min. J., 1894 (58), 270. 
t Trans. Am. Electrochem. Soc., 1910 (17), 327. 


desired concentration is produced. This is then entirely removed 
and a new portion of mercury introduced. By placing several com- 
partments in parallel and operating them successively, the cell be- 
comes practically continuous in its action. The cement cell consists 
of a shallow box having a decomposing compartment (A) and an 
oxidizing space (B). The graphitized carbon anodes (K), submerged 
in the brine in (A), are supported from the cell cover just above the 
surface of the mercury on the bottom of the chamber. The oxidizing 
chamber has inclined channels (P, P) forming a zig-zag path, leading 
to the pump-pit (Q). The current flows from (K) through the brine 
to the mercury, and out by the iron cathode (R). The sodium set 
free forms amalgam, which after a short interval of time is drawn off 
by the valve (E) into the oxidizing chamber (B), where the sodium is 
removed by water during the passage down the incline (P) to the 
pump-pit. The mercury, free from sodium, is elevated by a wheel 
pump (J) and returned to (A). The cycle of operations occupies about 
two minutes in each compartment. The cells are six feet square 
and take about 1400 amperes, with voltage approximately four. Chlo- 
rine of 98 per cent purity, with 2 per cent hydrogen, and pure caustic 
solution of any desired strength up to 40 per cent sodium hydroxide, 
is claimed. 

In Belt's apparatus * the mercury is moyed into the electrolysis 
chamber by the pressure exerted by the hydrogen evolved in the 
caustic compartment. 

In Rhodin's apparatus f the mercury is moved by the centrifugal 
action of the rotating earthenware bell which serves as the cover of 
the anode chamber. Neither the Bell nor the Rhodin cells have been 
commercially successful. 

The disadvantages of using mercury are the tendency of other 
metals, especially magnesium, derived from impurities in the brine, 
to accumulate in the amalgam and reduce its fluidity ; the high volt- 
ages (four or over) necessary ; the cost of the mercury itself, and of 
the installation in general. 

The " gravity " or " bell process " t (Fig. 61) employs a cell 
without diaphragm or mercury. An earthenware bell (B) is suspended 
in a tank containing brine. The anode (A) is near the top of the 

* Electrochem. Ind., 1903 (I), 505. 

t J. Soc. Chem. Ind., 1897, 745; 1900, 418; 1902, 449. 
Electrician, XL, 8. U. S. Pat. No. 608,300 (1898). 
I J. Soc. Chem. Ind., 1898, 1147; 1904, 545. 
Zeitschr. Elektrochem., 1901 (7), 581 ; 1904 (10), 317. 
Ber. 1908 (41), 1789; 1909 (42), 2904. 



liquid inside of the bell, and the cathode (K) is outside. New brine 
is constantly fed in by the pipe (C, C) and the caustic liquor overflows 
at (F), chlorine passing out at (D). The relatively large distance 

separating the electrodes raises 
the resistance to four volts. The 
cell is of small size, however, and 
an extensive plant is necessary 

FIG. 61. 

for a commercial enterprise. 

Electrolysis in fused baths is represented by the patents of Vautin, 
Heulin, and Acker. The Acker process * (Fig. 62) employs a bath 
of melted lead for the cathode, on which the fused salt rests ; a sodium- 
lead alloy forms and is decomposed by steam in a special chamber (C), 
producing anhydrous fused caustic directly. The cell is divided into 
three compartments ; in one. (A), the salt fused at about 850 C. is 
decomposed and the melted lead on the floor unites with the sodium. 

FIG. 62. 

The melted alloy flows into (B) and is ejected into (C) by a steam-jet 
from (F). The melted lead and fused sodium hydroxide separately 
gravity in (C), the lead flowing back to (A), and the caustic is drawn 
off. Each cell takes 8000 amperes at 6 to 7 volts. Much energy 
is consumed in fusing the salt, and by losses through radiation, con- 
duction, and resistance at the connections, and the up-keep expense 
is heavy. The only commercial trial of these cells was at Niagara 
Falls but the plant was destroyed by fire in 1907. 

The destructive action of caustic and chlorine on the diaphragms 
and other parts of the electrolytic apparatus, and the large size of 
the plant needed for a comparatively small output, are serious dis- 

* Trans. Am. Electrochem. Soc., I (1902), 165.- U. S. Pat. No. 649,565. 


advantages of electrolysis ; then, except in a few places where water- 
power is cheap, the electricity is generated with steam-engine and 
dynamo, a method of low efficiency, considering the fuel consumption. 

The electromotive force necessary to decompose salt is 2.3 volts ; 
but the resistance of the bath and polarization increase this to 3.5 or 
4 volts. One ampere of current yields, theoretically, 0.00292 pound 
of chlorine and 0.0033 pound of caustic soda per hour. If the 
efficiency is 80 per cent, one ampere yields 28.56 grams NaOH and 
25.2 grams Cl in 24 hours;* or, to make one kilo of NaOH in 24 
hours, the current must be 35 amperes. If a theoretical yield were 
obtained, the chlorine evolved would make about 100 pounds of 
bleaching powder for each 40 pounds of caustic produced. But the 
latter, which is in much greater demand than bleaching powder, can 
be made cheaply from ammonia soda ; this would seem to limit the 
electrolytic processes to supplying bleach and chlorates, while the 
caustic must be considered as a by-product. 

The caustic liquors produced by wet electrolysis in diaphragm 
processes are contaminated with salt and are dilute, requiring much 
evaporation. As the concentration of caustic in the electrolyte in- 
creases, there is increased carrying of current by the OH ions, with 
liberation of oxygen and formation of water. This causes such 
serious loss in strong solutions that the practical limit of concen- 
tration is about 12 or 15 per cent of NaOH. When mercury is used 
as cathode, strong, pure caustic is produced, with less consumption 
of fuel. 


By passing chlorine into a cold solution of sodium or potassium 
carbonate, a mixture of the chloride and hypochlorite of the alkali 
metal is formed. But if any excess of chlorine is introduced, the 
hypochlorite is decomposed into chloride and free hypochlorous 
acid (HOC1) : - 

1) K 2 C0 3 + H 2 + 2 Cl = KC1 + KOC1 -f H 2 O + CO 2 . 

2) K 2 C0 3 + H 2 + 4 Cl = 2 KC1 + CO 2 + 2 HOC1. 

This solution of hypochlorous acid is a powerful bleaching and oxi- 
dizing agent. It was first made about 1789, and brought into trade 
in France as a " bleach liquor " under the name of eau de Javelle, or 
eau de Labarraque. In 1798 or 1799 Charles Tennant took out a 
patent in England for a " bleach liquor " made by passing chlorine 

* Zeitschr. f. Elektrochem., 1895, 21. 


into "milk of lime," by which a solution of calcium chloride and hypo- 
chlorite was formed : 

2 Ca(OH) 2 + 4 Cl = CaCl 2 + Ca(OCl) 2 + 2 H 2 O. 

This bleach liquor is cheaper, stronger, and more convenient to use 
than bleaching powder (see below), but since it is unstable, evolving 
oxygen even when kept in a closed vessel in the dark, it is usually 
made only for immediate use. 

The tanks in which the milk of lime is treated with chlorine are 
provided with stirring apparatus ; the temperature must not rise 
much above 30 C., or chlorates are formed. A dilute chlorine may 
be used. The density of the solution obtained is about 8 Tw. 

Calcium carbonate suspended in water may also be employed for 
preparing bleach liquor : 

CaCO 3 + H 2 O + 4 Cl = CaCl 2 + CO 2 !- 2 HOC1. 

These liquors are chiefly used for bleaching vegetable fibres and 
for disinfectants. 

The absorption of chlorine in milk of lime soon led to trials of 
dry, slaked lime or calcium hydroxide for the same purpose. A dry 
bleaching powder, fairly stable and constant in strength, resulted; 
but its composition is not the same as that of the bleach liquor made 
from milk of lime. It was at first supposed that a direct combina- 
tion took place between the lime and chlorine, and that the powder 
was simply calcium hypochlorite [Ca(OCl) 2 ], so the name " chloride 
of lime " was given to it. Other investigations led to the view that 
it contained a mixture of calcium chloride and hypochlorite. But 
this was disproved by Lunge * and his students, who demonstrated 


the correctness of Odling'sf formula Ca . Hence it is an 

O Cl 

oxy chloride of calcium. When dissolved in water, this forms hypo- 
chlorite and chloride of calcium. 

For making bleaching powder, a pure, fat lime is desirable. It is 
slaked carefully, so that the resulting hydroxide contains about 24.5 
to 25.5 per cent of water, i.e. there should be a slight excess of water 
over that necessary to form calcium hydroxide. 

* Chemische Industrie, 1881, 289. Dingl. J., 237, 63. Annalen der Chemie, 
219, 129. Berichte d. deutsch, chem. Gesellschaft, 1887, 1474. Zeit. f. anorg. 
Chemie, II, 311. 

t Odling, Handbuch der Chemie, I, p. 59. 


The absorption chambers are brick, cast-iron, or lead, and are 
usually 6.5 feet high, and have about 200 square feet of floor area 
per ton of bleach made per week. Brick chambers are tarred inside 
to make them gas tight and to protect them from the chlorine ; large 
ones are usually made from lead, much like the vitriol chambers 
(p. 70), and may have a floor area of 30 by 100 feet. The slaked 
lime is sifted through screens with from 20 to 25 meshes per linear 
inch, as only the fine powder is suitable. This is spread three or 
four inches deep on the floor, and is furrowed with a special rake in 
order to assist the absorption by increasing the surface. The chlo- 
rine is introduced at the top of the chamber, and settling to the 
bottom because of its density, is at first rapidly absorbed by the lime. 

After a time the process goes on more slowly, and finally the gas 
enters under some pressure. Usually there are three or more chambers 
in a series, the strongest chlorine entering that containing the most 
nearly finished bleach, and passing out through that containing the 
fresh lime. The degree of absorption of chlorine is judged by the 
color of the gases seen through the glass " sights " in the chamber 
walls. The powder is turned over once or twice, and the treatment 
("gassing") continued until tests show that it contains from 36 to 
37 per cent of " available chlorine." If under strength (" weak "), 
after the third "gassing," it should be packed and sold for what it 
will bring, for further exposure will cause the formation of chlorate 
and chloride with loss of strength. 

During the absorption considerable heat is generated ; for strong 
powder the temperature should not exceed 40 to 46 C.* The chlo- 
rine should be admitted in a very slow stream, and should be con- 
centrated, dry, and free from hydrochloric or carbonic acids. When 
dilute (as from Deacon's apparatus), a large, special chamber pro- 
vided with numerous shelves, on which the slaked lime is spread 
to secure a greater absorbing surface, is employed ; or the apparatus 
shown in Fig. 63 is used. The yield from 100 pounds of good lime 
is about 150 pounds. 

Mechanical apparatus (Fig. 63) for the absorption of the chlorine 
is much used. Several horizontal cast-iron cylinders (A), set one 
above the other, each contains a rotating shaft carrying blades which 
act as conveyers ; the shafts in the several cylinders are all driven at 
the same speed by a system of gears (B). Slaked lime is fed con- 
tinuously in a small stream to the upper cylinder through (H), and 
is carried by the blades to the opposite end of the cylinder, where 

* Lunge and Schappi, Dingl. J., 237, 63. 



it drops through the opening (D) into the next, and so on to the 
bottom. Chlorine enters the lowest cylinder at (C), passes over the 

surface of the lime, ascends 
through (D) to the next 
higher cylinder, and thence 
up through each in succes- 
sion, the unabsorbed gas 
finally escaping at the top 
through (E). The bleach- 
ing powder thus formed 
collects in (F), from which 
it is dropped directly into 
the casks for packing. 

Bleaching powder is a 
yellowish white substance, 
which should be perfectly 

dry and free from lumps. On exposure to the air, it absorbs mois- 
ture and carbon dioxide, giving off hypochlorous acid, the evolution 
of which gives bleach its peculiar odor. Good samples contain about 
36 per cent " available chlorine." Its chief use is for bleaching vege- 
table fibres for the textile and paper industries. 

In order to liberate the chlorine for bleaching purposes, the powder 
is usually decomposed by a mineral acid, thus : the fibre, having been 
saturated with the bleaching powder solution, is passed into a dilute 
acid bath, where the hypochlorite is decomposed and the chlorine set 
free. The nascent chlorine combines with the hydrogen of the water, 
liberating nascent oxygen, which, in turn, destroys the organic coloring 
matter in the fibre. 

Bleaching powder, obtained from a cheaper base and giving higher 
chlorine efficiency, has replaced hypochlorites made from alkali car- 
bonates and chlorine (p. 131). Free hypochlorous acid oxidizes 
hypochlorites to chlorates (NaCIO + 2 HC1O = NaClO 3 + 2 HC1), but 
in alkaline solution, where no free acid exists, they are relatively 
stable. The action of chlorine on carbonates liberates hypochlorous 
acid, which thus destroys hypochlorite. The use of alkali hydroxides 
avoids this loss (2 OH' + C1 2 = Cl~ + CIO" + H 2 O). As liquid 
chlorine is now a commercial article, hypochlorite solutions made by 
absorbing the gas in caustic liquor are much used ; this has the advan- 
tage over lime that it is easily washed from the goods and forms no 
insoluble soaps. 


Hypochlorites in dilute solution are strong oxidizing agents on 
organic matter; chlorates do not have this action, hence only the 
hypochlorite of the solution is " available " for bleaching purposes, 
and it is only the oxygen of the hypochlorite which is active. The 
chlorine equivalent of this hypochlorite oxygen is called the " avail- 
able chlorine." 


The chlorates are stable salts and are always made by decomposi- 
tion of hypochlorites in hot solution (see above). Formerly chemical 
chlorine was run into calcium or magnesium hydroxide (Liebig's 
process). From the reactions it is seen that five times as much chloride 
is formed as chlorate. The chlorate solution was double decomposed 
with potassium salts, with precipitation of potassium chlorate, it 
being less soluble in the calcium or magnesium chloride solution. A 
large part of the product was lost, however, in the mother-liquors. 

These methods are now entirely superseded by the production 
of the chlorine and alkali by electrolysis of alkali chloride solutions, 
no diaphragm being used, thus allowing anode and cathode products to 
interact. The chloride regenerated by the above reactions is re- 
electrolyzed, thus enabling the conversion of all the chloride to chlo- 
rate. Since the hypochlorite is subject to reduction by the nascent 
hydrogen at the cathode forming water according to the reaction, 

2 H + CIO" = Cr + H 2 O, 

and to deposition at the anode with loss of oxygen, all representing 
lost efficiency, the conversion to chlorate is made as rapid as pos- 
sible by keeping the solution hot and slightly acid. The current 
efficiency can be made from 85 to 90 per cent. The voltage 
required for the deposition of chlorate ion is much higher than 
that for chlorine or hydroxyl ion, hence, so long as chloride re- 
mains in the bath, no chlorate deposits, but when all chloride is oxi- 
dized, chlorate deposits, best at high-current densities, with formation 
of perchlorate, C1O 3 ~ + O = ClOr, the oxygen coming from anodic 
deposition of chlorate. Perchlorates, especially those of potassium 
and ammonium, are now made in this way, and are largely used for 
explosives which possess great stability against shock. 

The crude chlorate is purified by recrystallizing from water, and 
the crystals are drained and washed in a centrifugal machine, and 
may be sold as coarse crystals ; or they are ground to fine powder in 
buhrstone mills, care being taken that no organic matter, dirt, or 


metal (iron, etc.) gets into the mill, lest an explosion result. No 
fire should be permitted in the building, and heating should be by 
steam, and lighting by electricity. The grinding mill should be 
at a distance from the main works. 

Sodium chlorate is much more soluble and is more difficult to crys- 
tallize than potassium chlorate, but it is, however, made in large 

Persulphates of potassium and ammonium are made by electro- 
lyzing at very low temperatures strong solutions of the acid sul- 
phates or acid carbonates with high current densities on smooth 
platinum anodes. The anions KSO~ 4 and KCO~ 3 combine to form 
persulphate (K^Os) or percarbonate (K 2 C 2 O 6 ), which crystallize as 
moderately stable salts. In solution they are active oxidizing agents, 
and are used to a limited extent for bleaching and oxidizing purposes. 


Die Fabrication von chlorsaurem Kali und anderen Chloraten. Dr. 

Conrad W. Jurisch, Berlin, 1888. (R. Gaertner.) 
Die Darstellung von Chlor u. Salzsaure, unabhangig von der LeBlanc 

Soda Industrie. Dr. N. Caro, Berlin, 1893. (R. Oppenheimer. ) 
Sulphuric Acid and Alkali. G. Lunge, 3d ed., Vol. Ill, New York, 1911. 

(Van Nostrand Co.) 
J. Soc. Chem. Ind. : 1883, 103, Ferdinand Hurter. 

1885, 525, W. Weldon. 1887, 248, C. Longuet Higgins. 

1887, 775, James Dewar. 1896, 713, Ludwig Mond. 

Electro-Chemistry. M. Leblanc, translated by W. R. Whitney and 

J. W. Brown, New York, 1907. (Macmillan & Co.) 
Elements of Electro-Chemistry. Lupke. 
Die Fabrikation der Bleichmaterialien. V. Holbliug, Berlin, 1902. 

Grundriss der reinen und angewandten Elektrochemie. P. Ferchlaud, 

Halle a. S., 1903. (W. Knapp.) 

Elektrochemie wassriger Losungen. F. Forster, Leipzig, 1905. 
Applied Electrochemistry. By M. deKay Thompson, New York, 1911. 

(Macmillian Co.) 


The manufacture of nitric acid is often carried on in conjunction 
with sulphuric acid making, especially in those plants where the liquid 
acid is used in the lead chamber process. Large quantities, however, 
are produced for general manufacturing purposes. 

Practically all nitric acid is made by treating sodium nitrate 
(p. 145) with sulphuric acid in cast-iron retorts. The reactions are 
as follows : 

1) NaNO 3 + H 2 SO 4 = NaHSO 4 + HNO 3 . 

2) 2 NaNO 3 + H 2 SO 4 = Na 2 SO 4 + 2 HNO 3 . 

But it is now thought that a sodium acid sulphate* is formed and 
reacts with some of the sodium nitrate, 

3) NaNO 3 + 2 H 2 SO 4 = NaH 3 (SO 4 ) 2 + HNO 3 . 

4) NaH 3 (SO 4 ) 2 + NaN0 3 = 2 NaHSO 4 + HNO 3 . 

In practice the quantities of material used do not correspond with 
either of these equations, but the charge is so regulated that a mixture 
of acid and neutral sulphates of sodium, which remains liquid at the 
temperature employed, is left in the retort. If reaction (1) were 
followed, too much sulphuric acid would be used for profitable work- 
ing, except in soda works, where the resulting acid sulphate is used 
in the salt-cake furnace. If reaction (2) is carried out, the tempera- 
ture must be high, and the nitric acid is partly decomposed by the 
heat, before it can escape from the retort, causing smaller yield and 
a product discolored by the nitrogen oxides produced. 

The sulphuric acid employed is usually that from the lead pan 
evaporation (sp. gr. 1.70), but for strong nitric acid the sulphuric 
acid should be of 92 per cent strength, and the temperature kept as 
low as possible during the distillation. The sodium nitrate used is 
purified Chili saltpetre having 96 to 99 per cent NaNO 3 when dry. 
Chlorides should not be present because of the decomposition re- 
sulting (p. 141). The size of charge ranges from 500 to 1200 pounds 
or more of nitrate, and 60 Be. sulphuric acid, amounting to 20 or 30 
per cent excess over the theoretical quantity. 

The apparatus (Fig. 64) commonly used consists of a horizontal 
cast-iron cylinder or retort (A) about six feet long by four or five feet 

* J. Am. Chem. Soc., 1901 (23), 489. 



diameter, set in a furnace in such a way that the flame plays over 
its entire surface, heating all parts equally hot. Cast-iron is but 
little attacked by concentrated nitric acid or its vapors, and it is 

important that the retort be hot enough 
in all parts to prevent any condensation 
of acid. Retorts having exposed ends 
made of sand-stone slabs, or vitrified 
brick laid in cement, are sometimes used. 
The charge of nitre is introduced by a 
door in the end, or side, of the retort, 
and the sulphuric acid is run in by a 
pipe (E). After the reaction the fused 
residue is run off by (D) and on cooling 
FlG - 64 - forms the so-called " nitre cake." 

The acid vapors escaping from the retort are condensed in earth- 
enware pipes or worms, surrounded by water, or irf a series of Woulfe 
bottles (bombonnes) (Fig. 44) with an absorbing tower at the end to 
catch the fumes escaping from the bottles. The dilute acid from the 
tower may or may not flow through the series of bottles in a direction 
opposite to the movement of the acid vapors. 

The most concentrated acid condenses in the first two or three 
bottles but is contaminated with sulphuric acid; the last of the 
series contains dilute acid, which is contaminated with chlorine; 
in the middle bottles is pure acid of moderate strength. Owing to 
more or less reduction of the nitric acid in the retort, the condensed 
acid has a yellow or red color, due to absorbed nitrous vapors. For a 
commercial acid, these must be removed by " bleaching " ; the acid 
is heated to about 90 C., and air blown in 
which carries away the nitrogen oxides to an 
absorbing tower for recovery. 

Guttmann's apparatus * is much used. The 
large cast-iron retort (Fig. 65) is made in three 
pieces and is entirely surrounded by the flames 
from the grate. The retort gases pass into a 
system (Fig. 66) of vertical earthenware pipes 
(A, A) having very thin walls and joined at the 
top by bends, while they open at the bottom 
into a nearly horizontal collecting pipe (B, B) 
which is divided into sections by diaphragms. The sections are con- 
nected by U-tubes passing under the partitions. The diaphragms 
* J. Soc. Chem. Ind., 1893, 203. 

FIG. 65. 



force the acid vapors to pass up one pipe and down the next, in 
order to go through the system. The thin walls (8 mm.) of the 
vertical pipes allow efficient and rapid cooling by the cold water 
in the tank, and the vapors are quickly condensed. Air at 80 
C. is injected from (F) into the outlet pipe (D), where it con- 
verts some of the nitrous vapors into nitric acid, increasing the 
yield. The uncondensed nitrous vapors pass into the Lunge-Rohr- 
mann plate tower (E), where they are absorbed in sulphuric acid or 

FIG. 66. 

water. If the vapors remain in the retort too long, part of the acid 
is decomposed and nitrogen peroxide is formed and absorbed by the 
condensed acid, to which it imparts a red color. Since there is good 
draught in this apparatus, the vapors are drawn out of the retort soon 
after they are evolved, and are at once condensed ; very little peroxide 
is formed, and a light-colored, concentrated acid is obtained. It is 
claimed that 40 Be. acid, requiring no " bleaching," is made, and 
with water-cooled pipes, 98 per cent of the theoretical yield is 
obtained as concentrated acid, while 2 per cent condenses in the 
plate tower. 

Hart's tube condenser * (Fig. 67) for nitric acid is made of glass 
and earthenware tubes, and is placed above the brick arch over the 
retort, thus occupying but little floor space. The vapor from the 
retort (A) passes into the pot (B) and thence through the vertical 
earthenware tube (C). From (C) to (D) extend a number of glass 
tubes, inclined slightly towards (C), and cooled by jets of water from 
the perforated pipe (E, E). From (D) the uncondensed vapors pass 
* J. Soc. Chem. Ind., 1894, 1197. J. Am. Chem. Soc., 1895 (17), 576. 



to a plate tower. The acid condensed in the glass tubes flows back 
into (C), and then into (B), thus coming into contact with the hot 
vapors from the retort. This heats the acid so hot that the nitrous 

vapors are driven out and 
the light-colored acid is run 
off by the U-tube (F). In 
this apparatus the acid is 
condensed very quickly and 
little peroxide is formed. 
Frothing in the retort gives 
no trouble, as any overflow 
is caught in (B) and is easily 
removed. The water flow 
from (E) can be so regulated 
that nearly all of it is evap- 
orated on the surface of the 
glass tubes, giving great cool- 
ing effect with small con- 
sumption of water ; any ex- 
cess of condenser water is 
caught in (G). The chief 
repairs are of broken tubes, which are cheaply and easily replaced. 

Valentiner's process * consists in distilling the mixture of sul- 
phuric acid and nitre in a vacuum of from 60 to 65 cm. of mercury, the 
temperature being about 85 C., but finally rising to 160 C. There 
is very little decomposition of the nitric acid, and a large yield of con- 
centrated, light-colored acid is obtained. The charge is about 1000 kg. 
of Chili nitre, with the requisite amount of 60 Be. sulphuric acid, 
and provision is made for frothing. The acid gases are condensed in 
an earthenware worm, cooled with water. The uncondensed vapors 
pass through milk of lime, and then to a bronze vacuum pump. The 
yield of acid is claimed to reach 98 per cent of the theoretical. 

The Rhenania process f is based upon the reactions between fused 
sodium bi sulphate and sulphuric acid, forming the so-called poly- 
sulphate, which in turn reacts with the sodium nitrate : 

1) NaHS0 4 + H 2 SO 4 = NaH 3 (SO 4 ) 2 . 

2) NaH3(SO 4 )2 + NaNO 3 = 2 NaHSO 4 + HNO 3 . 

FIG. 67. 

* J. Soc. Chem. Ind., 1893, 155 ; 1896, 36 ; 1899, 492, 1122 ; 1901, 544. 

Zeit. angew. Chem., 1899, 269, 1003. Chem. Zeit., 1895, 118 ; 1897, 511. 
t Chem. Ind., 1901, 189, 544, 896, 1189. J. Soc. Chem. Ind., 1902, 173. 
J. Am. Chem. Soc., 1901 (23) 489. 



The apparatus (Fig. 68) consists of three iron retorts; two (A) 
heated to 200 C. by the waste heat from the third (B) at 300 C. 
Sodium nitrate and fused poly- 
sulphate are introduced into (A), 
where reaction (2) runs for a con- 
siderable time, and is then com- 
pleted by drawing the charge into 
the hotter retort (B), containing 
some fused bisulphate from the 
previous operation. Here reac- 
tion (2) is completed and the last 
of the nitric acid driven out. 
Most of the bisulphate is then FlG - 68> 

drawn from (B) and part of it heated with 60 Be. sulphuric acid 
in a pan (C) to form polysulphate for the next charge. Since 
the reaction in (A) is slow, while the final one in (B) is rapid, the 
two retorts (A) are connected with one retort (B), and by running 
alternate charges in the former to be finished in (B), the process be- 
comes practically continuous. Retort (A) yields concentrated acid, 
but that from (B) is rather dilute. 

The strength of the nitric acid produced depends upon the strength 
of the sulphuric acid, on the temperature of the retort, and on the 
purity of the sodium nitrate. With sulphuric acid of 1.71 sp. gr. 
the nitric acid varies from 1.38 to 1.42 sp. gr. (40 to 42 Be.). If 
the sodium nitrate contains chlorides, some of the nitric acid is de- 
composed by the hydrochloric acid produced : 

HN0 3 + HC1 = H 2 + NO 2 + Cl. 

For chemically pure acid, pure materials are used ; but the common 
acid may be purified by treating with silver and barium nitrates and 
redistilling. Concentrated acid cannot be distilled without some 
decomposition, and the product must be " bleached " by heating 
and blowing in pure air. 

Fuming nitric acid is a solution of nitrogen peroxide in concen- 
trated nitric acid. It is red in color and has a specific gravity of 1.55 
to 1.62. To make this, perfectly dry sodium nitrate and oil of vitriol 
(1.84 sp. gr.) are used. The reaction is carried so far that neutral 
sodium sulphate is formed : 

2 NaN0 3 + 2 NaHSO 4 = 2 Na 2 SO 4 + 2 NO 2 + H 2 O + O. 

The nitrogen peroxide dissolves in the nitric acid to form the fuming 
acid. A little starch may be added to assist in the reduction. An 


impure fuming acid is made by distilling a mixture of concentrated 
nitric and sulphuric acids. 

Priestley and Cavendish observed that nitric acid is formed by the 
action of electric sparks on damp air. This results from the direct 
union of oxygen and nitrogen at high temperature, to form nitric oxide : 
N + O ^t NO. The reaction is endothermic and is accelerated by 
high temperatures, with larger yields of nitric oxide. But for each 
temperature, an equilibrium is reached, corresponding to a definite 
percentage of nitric oxide, and the yield is not increased by longer 
heating at that temperature. With air at 2000 C., this equilibrium 
is reached almost instantly, but at 1500 C. the reaction is slow. 
According to Nernst * only 0.1 per cent by volume of nitric oxide is 
produced at 1500 ; 0.37 per cent at 1811, and 0.97 per cent at 2195. 
At higher temperatures larger percentages are produced, until at 3327 
about 5 per cent of nitric oxide should be formed. But as the reaction 
is reversible, there is much decomposition of the product during its 
cooling, so the yield is decreased. Below 1500 C., decomposition of 
the product is slight, so very rapid cooling to this temperature is 
essential. As the electric arc is the most feasible method of attaining 
high temperatures, this method of producing nitric acid direct from 
the air has attracted considerable attention. f 

The first technical attempt at the electrical fixation of atmospheric 
nitrogen was the Bradley and Lovejoy process. { A spark at about 
10,000 volts is jumped a short distance through the air, and the arc 
formed is immediately broken. The machine makes about 414,000 
arcs per minute, and the nitrogen oxides in the air leaving the appa- 
ratus are condensed in towers to form nitric and nitrous acids. The 
process failed industrially. 

The Birkeland and Eyde process has been more successful, al- 
though mainly as applied to making nitrates rather than the acid. 
It is based on the fact that in a magnetic field the electric arc is de- 
flected to one side. The furnace (Fig. 69) is a narrow, disk-shaped- 
vertical box with fire-brick lining. Two copper electrodes (E), 
internally cooled by water, project horizontally into the furnace oppo- 
site to each other, and one-third of an inch apart. Around the elec- 

* Zeitschr. anorg. Chem., 1905 (45), 126; 1906 (49), 213. 

t J. Soc. Chem. Ind., 1906, 567 ; 1915, 113. 

j J. Soc. Chem. Ind., 1902, 1138. Electrician, 1902, 684. U. S. Pat. Nos. 
709,867, and 709,868. 

Zeitschr. angew. Chem., 1905, 217. Chem., Ind., 1905, 699 ; 1915, 114. Elec- 
trochem. Met. Ind., 1904, 399 ; 1906, 126. 



FIG. 69. 


trodes are the coils of an electromagnet, by which the arc is deflected 
until it breaks and a new arc forms, to go through the same cycle. 
The speed of formation, deflection, and breaking 
of the arc is so rapid that about 700 arcs per second 
are formed, and a " disk of flame " is produced. 
Air enters the oven by narrow channels (A), and 
encounters the arcs in (B), leaving the furnace by 
the flues (C) mixed with about 2 per cent of nitric 
oxide. The nitric oxide is oxidized rapidly to 
peroxide and the gases go to quartz-packed ab- 
sorbing towers, fed with water on the counter- 
current principle. Nitric acid of 50 per cent 
strength is obtained, which is mostly used to make 
calcium nitrate for fertilizer. Furnaces of 1000 to 3000 kilowatts 
WATER .NLET capacity are used, and a yield of 500 to 

600 kilograms of anhydrous nitric acid 
per kilowatt-year is claimed. 

The Schoenherr process * exploited 
by the Badische Anilin u. Sodafabrik 
uses the apparatus shown in Fig. 70. 
The arc is sprung inside of a long narrow 
iron tube, having in the lower end an 
insulated iron electrode (E), which is 
pushed forward as needed through a 
copper water-jacket. The air enters at 
(A), passes through concentric pipes sur- 
rounding the reaction tube, and enters 
the latter near the electrode (E) by 
means of the tangential inlets in the 
movable sleeve (S). The air is thus 
given a spiral, rotary motion through 
the tube. The upper end of the reaction 
tube is water-jacketed for one-third its 
length and the tube itself forms the other 
electrode. The arc is started by moving 
the bar (Z) to short-circuit from the 
electrode (E) to the reaction tube; the 
arc thus starting in the centre of the 
FIO. 70. whirling current of air, one end of it is 

* Zeitschr. angew. Chem., 1338, 1633. J. Soc. Chem. Ind., 1915, 115. Met 
Chem. Engineering, 1909, 245. 



driven along the tube wall until it reaches the water-cooled section, 
where it again strikes the tube wall. The arc is thus extended to 5 or 
7 metres length, and the air surrounding it is in contact with it for rel- 
atively long time. The hot gases pass out of the cooled end of the 
reaction tube and enter an annular space surrounding the inlet pipe, 
to which they impart some of their heat, for pre-heating the incom- 
ing air. Furnaces taking 1000 H. P. are in use and yields of 140 grams 
HNO 3 per kilowatt-hour are claimed.* Only 3 per cent of the total 
energy supplied is used for the production of nitric oxide, which is 
oxidized to peroxide by the excess air present, when cooled below 600 
C. The absorption of the mixed nitric oxide and peroxide gases 
from the furnace is troublesome, since the reaction 

3 N0 2 + H 2 = 2 HNO 3 + NO 

regenerates much nitric oxide. If this, with air, is led into hot soda- 
liquor, sodium nitrite is obtained : 

2 NO + O = NO + NOa. 

2 NaOH + NO + N0 2 = 2 NaNO 2 + H 2 O, or 

2 NaOH + 4 NO = N 2 O + 2 NaNO 2 + H 2 O. 

This method for nitrite has replaced the old reduction process from 
sodium nitrate by metallic lead. Milk of lime yields calcium nitrite, 
which is used as a fertilizer under the name " air-saltpetre." When 
the dilute nitric acid from the water absorption is mixed with the 
calcium nitrite liquor and evaporated to dryness, some nitrous acid 
escapes and calcium nitrate is obtained, which is sold as " Norway 
saltpetre," for fertilizer. 

The Pauling process f uses an apparatus shown in diagram by 
Fig. 71. The arc formed between the diverging electrodes (A, A) 
is elongated to about one metre, into the fan- 
shaped space between (A, A) by a blast of air 
from the nozzle (E). Each furnace is said to take 
400 kilowatts at 4000 volts for treating 600 cubic 

FIG. 71. 

metres of air. The gases after oxidation to per- 
oxide are absorbed in water to produce nitric acid., The yield is said 
to be 60 grams of anhydrous -acid per kilowatt-hour. 

Nitric acid is largely used in making explosives ; for parting gold 
and silver ; in the manufacture of coal-tar dyes and other products ; 

* Zeitschr. angew. Chem., 1909, 1174. 

f J. Soc. Chem. Ind. 1907, 1204; 1909, 1317; 1915, 115. 

J. Ind. Eng. Chem., 1914, 68. 


as a " pickling liquor " for cleaning metal ; for various etching 
processes; and in making metallic nitrates. The pure acid is a 
colorless liquid, boiling at 86 C., but with decomposition. It also 
decomposes on exposure to strong light and becomes yellow (NOfe). 
Ordinary commercial acid (1.42 sp. gr.) distils at 123 C., contains 
68 to 69 per cent HNO 3 , and concentrated acid of 1.50 sp. gr. con- 
tains about 94 per cent HNOs. 


The most important nitrates are those of sodium and potassium, 
but ammonium, lead, iron, silver, strontium, and barium nitrates 
are used to some extent in the arts. 

Sodium nitrate,* also called Chili saltpetre, is found in natural 
deposits in desert regions along the west coast of South America, 
especially near the boundary lines between Peru, Chili, and Bolivia, 
in latitude 20 to 26 S. The territory is now chiefly owned by 
Chili. The deposits extend about 220 miles in length, and average 
about 2 miles in width. 

The crude nitrate, called " caliche," varies from yellowish white 
to brown or gray, and contains from 20 to 55 per cent NaNO 3 ; it 
forms beds about 5 feet thick, lying near the surface, but usually 
covered by a conglomerate of rock debris, cemented together by salt 
and gypsum. The region is rainless, and water and fuel, being very 
scarce, are used as economically as possible in refining the crude 
ore. The caliche is crushed and boiled with water in tanks heated 
by steam coils, until the liquor reaches a density of 110 Tw., when 
it is run off to crystallize. The mother-liquor retains most of the 
chloride, iodide, and iodate of sodium and magnesium, together with 
about 20 per cent of the nitrate. Hence the liquors are diluted with 
the wash water from the residue, and used again to lixiviate another 
portion of caliche. But after two or three repetitions of this process 
the mother-liquor is too contaminated for further use. It is then 
run off and treated for the recovery of the iodine (p. 252) which it 
contains. The residue from the lixiviation contains some nitrate, 
and is washed with fresh water, yielding a weak solution, which is 
used to dilute the mother-liquors before using them for leaching. 
The sodium nitrate crystals are drained or " centriffed " and dried 
in the sun. They are then packed and shipped as crude Chili salt- 

* J. Soc. Chem. Ind., 1890, 664 ; 1893, 128. 


petre, containing from 94 to 98 per cent of NaNOs. For many pur- 
poses this is purified by recrystallization. 

Deposits of sodium nitrate have been found recently in Upper 
Egypt and in the trans-Caspian region, but these have not been much 
developed as yet, and nearly all the world's supply comes from Chili. 

The formation of these beds is attributed to the decomposition 
of sea-plants under such conditions of temperature and humidity 
that the ammonia produced was converted into nitrate by the action 
of the nitrifying bacillus, an organism found in the soil. The region 
being rainless, the sodium nitrate was not washed away. 

Potassium nitrate, or saltpetre, is derived from three sources : 

1. Natural nitrate beds, formed by the decomposition of organic 
matter in warm, damp climates. 

2. Artificial nitrate beds, prepared especially for the purpose. 

3. The decomposition of sodium nitrate by potassium chloride. 
In many tropical countries, especially in India, Persia, and Egypt, 

native deposits of potassium nitrate are found impregnating the 
earth in the neighborhood of large cities and towns. This forma- 
tion is due to the action of the nitrifying bacteria, and is not strictly 
an oxidation process. The deposits are continually forming, a 
white efflorescence appearing on the surface of the ground. This 
is scraped up, lixiviated with water, and the clarified solution evap- 
orated directly, to crystallize the nitre. But all the calcium nitrate 
in the mother-liquors is thus lost. By adding potash obtained from 
wood ashes the calcium nitrate is decomposed, and a larger yield of 
nitre is obtained. 

The artificial production of saltpetre in beds of decaying organic 
matter is now of slight importance, though formerly largely prac- 
tised in Sweden, Switzerland, and France when nitre was collected 
as a part of each farmer's tax. By this process putrefying organic 
matter is mixed with old mortar, or with porous earth containing 
calcium carbonate and wood ashes, and the pile allowed to stand for 
some months, being occasionally moistened with the liquid drainage 
from stables. The nitrifying organisms soon impregnate the mass 
with nitrates of calcium, potassium, and magnesium. On leaching, 
these go into solution ; when boiled with wood ashes, the calcium 
and magnesium are precipitated as carbonates, while the clarified 
liquor yields potassium nitrate on concentrating. The solution is 
clarified by adding a little glue, which combines with the impurities, 
forming a scum, which is removed by skimming. 

Potassium nitrate, made by double decomposition of sodium 


nitrate with potassium chloride, is now the most important from a 
commercial standpoint. The reaction is very simple : 

NaNO 3 + KC1 = NaCl + KNO 3 . 

Commercial potassium chloride, containing about 80 per cent KC1, is 
dissolved in water in cast-iron, copper, or lead lined wood tanks hold- 
ing 500 to 600 gallons. When the hot solution has a density of 
about 40 to 42 Tw. (1.20 to -1.21 sp. gr.), sodium nitrate containing 
95 per cent NaNO 3 is added, and the boiling mixture well stirred for an 
hour. On evaporation, the common salt, being less soluble than the 
nitrate, precipitates, and as much as possible of it is " fished " out, 
the concentration being continued until the density of the solution 
is 100 Tw. (1.50 sp. gr.). The liquid is allowed to stand a short 
time to settle, and then, while still hot, is drawn from the sediment 
into crystallizing tanks, where it is actively stirred while cooling. 
This causes the separation of the nitre as " crystal meal ", which is 
washed with a saturated solution of potassium nitrate (or often with 
cold water) to remove the mother-liquor and remaining sodium 
chloride. The wash waters and mother-liquors are used to dissolve 
the next lot of potassium chloride. One or two recrystallizations 
free the potassium nitrate from all but a trace of chloride. 

When the potassium chloride contains some magnesium chloride, 
it is best to precipitate the magnesium by soda-ash before adding 
the sodium nitrate, since traces of magnesium chloride may other- 
wise remain in the product. This salt, being deliquescent, may 
cause the nitrate to become wet on exposure. 

The chief uses of potassium nitrate are for making gunpowder 
and explosives, in matches, in pyrotechnics, in assaying, in metal- 
lurgical and analytical operations, and for curing meat. 

Ammonium nitrate is now used to a considerable extent in the 
manufacture of certain " flameless " explosives, and also, in a less de- 
gree, for making nitrous oxide (" laughing gas "). It is usually made 
by neutralizing nitric acid with ammonia. Attempts to produce it 
by double decomposition of sodium nitrate with ammonium salts 
result in incomplete reactions, and some sodium nitrate remains un- 

Lead nitrate is generally made by dissolving litharge (PbO) in 
hot dilute nitric acid. After filtering, the solution is concentrated 
to a density of 100 Tw. (1.50 sp. gr.) and allowed to crystallize. 

It is used in dyeing and calico printing, for the manufacture of 
certain orange and yellow pigments (chrome yellows), for some 


explosives, and in some kinds of matches. It is important in that 
it furnishes a moderately soluble lead salt. 

Ferric nitrate (nitrate of iron) is generally made by dissolving 
scrap iron in nitric acid of 1.30 sp. gr. The reaction is as follows : 

2 Fe + 8 HN0 3 = 2 Fe(NO 3 ) 3 + 2 NO + 4 H 2 O. 

By concentrating the solution, colorless crystals, containing six or 
nine molecules of crystal water, are obtained. 

The aqueous solution will dissolve ferric hydroxide, and this basic 
solution is much used in textile coloring. By using an excess of 
iron, and permitting the reaction to continue slowly after all the acid 
has been acted upon, a precipitate of insoluble basic ferric nitrate 
ultimately forms. The solution obtained in this way is of a red- 
brown color and indefinite composition. It is chiefly used for blacks 
in silk dyeing, and for iron-buff on cotton. 

Ferrous nitrate is prepared by dissolving iron in cold dilute nitric 
acid (1.10 sp. gr.). But a considerable amount of ammonium nitrate 
is also formed in the solution, according to the reaction : 

4 Fe + 10 HNO 3 = 4 Fe(NO 3 ) 2 + NH 4 NO 3 + 3 H 2 O. 

This solution is very unstable and decomposes when heated even 
slightly, forming basic ferric nitrate and liberating nitric oxide. 

To prepare a pure ferrous nitrate, decomposition of a ferrous 
sulphate solution by barium or lead nitrate is employed : 

FeSO 4 + Ba(N0 3 ) 2 = BaSO 4 + Fe(NO 3 ) 2 . 

The solution is filtered or decanted from the precipitated barium 

There is a preparation sold as " nitrate of iron " (probably so 
called because some nitric acid is used in making it), which is not 
a nitrate, but a basic ferric sulphate and sulphate-nitrate solution. 
A solution of ferrous sulphate (copperas) is oxidized by nitric acid, 
according to the following equations : 

1) 6 FeSO 4 + 2 HN0 3 + 2 H 2 O = 3 Fe 2 (SO 4 ) 2 (OH) 2 + 2 NO. 

2) 6 FeS0 4 + 5 HNO 3 = 3 Fe 2 (SO 4 ) 2 ' (NO 3 ) ' (OH) + 2 NO + H 2 O. 

3) 6 FeS0 4 + 8 HNO 3 = 3 Fe 2 (SO 4 ) 2 (NO 3 ) 2 + 2 NO + 4 H 2 O. 

4) 12 FeSO 4 + 3 H 2 SO 4 + 4 HNO 3 = 3 Fe 4 (SO 4 ) 5 "' (OH) 2 + 4 NO 

+ 2 H 2 O. 

Equation (4) gives the best product. 

The solution of basic ferric sulphate and sulphate-nitrates is a 
dark brown-red liquid, and is much used in silk dyeing. It is only 


mentioned here because of the frequent confusion of names in the 
commercial article. 

Silver nitrate is made by dissolving the metal in dilute nitric 

acid : ~ 6 Ag + 8 HN0 3 = 6 AgN0 3 + 4 H 2 O + 2 NO. 

If the silver contains copper, the resulting solution of nitrates is 
evaporated to dryness and then heated cautiously to about 250 C., 
at which temperature the copper nitrate is decomposed into copper 
oxide, nitric oxide, and oxygen, while the silver salt is not altered. 
By extracting the residue with water, the silver nitrate is dissolved, 
leaving the copper oxide. The solution is then evaporated to crys- 
tallize the silver nitrate. 

The salt fuses unchanged at 225 C., but decomposes if heated 
nearly to redness ; it is cast in small sticks, and is used in medicine 
for a cautery, under the name of lunar caustic. Silver nitrate has a 
very corrosive action on organic matter. It is largely used in pho- 
tography, and to a lesser degree in pharmacy, in the manufacture of 
mirrors, in preparing " indelible inks," and as a chemical reagent. 

Barium nitrate is made by dissolving the native carbonate (wither- 
ite) in hot, dilute nitric acid ; or it may be prepared by decompos- 
ing a concentrated solution (32 Be.) of barium chloride, by the 
addition of sodium nitrate, the less soluble barium nitrate precipi- 
tating. The salt is purified by recrystallization. It is chiefly used 
for producing " green fire " in pyrotechnics and for making barium 
peroxide (BaO 2 ) (p. 272). It is also used as an oxidizing material in 
certain explosives. 

Strontium nitrate is made by dissolving the native carbonate 
(strontianite) in hot nitric acid. Its chief use is for " red fire " in 
pyrotechnics. REFERENCES 

Berichte iiber die Entwickelung der chemischen Industrie, u. s. w. A. W. 

Hofmann, 1877. (Vieweg, Braunschweig.) 

Sulphuric Acid and Alkali. G. Lunge. 3d ed., Vol. I, 1903. (London.) 
The Manufacture of Explosives. Oscar Guttmann. (Nitric acid and 

Der Chilisalpeter und Zukunft der Salpeterindustrie. H. Polakowsky. 

Directorium der landwirthschaftl. Hauptgenossenschaft zu Berlin. 

Berlin, 1893. 
Die technische Ausniitzung des atmospharischen Stickstoffes. E. Donath 

u. K. Frenzel, Leipzig, 1907. 

Zeitschrift f. angewandte Chemie., 1893, 37. Oscar Guttmann. 
Journal American Chemical Society, 1896, 576. Edward Hart. 
J. Soc. Chem. Ind., 1893, 128. J. Buchanan. (Sodium nitrate in Chili.) 

1893, 203. Guttmann. (Nitric acid.) 1905, 924. 
Utilization of Atmospheric Nitrogen. By Thomas H. Norton. Bull. 

No. 52, Special Agents Series, Department of Commerce and Labor. 

Bureau of Manufactures. Washington, D.C., 1912. 


The destructive distillation of organic matter containing nitrogen 
yields more or less ammonia, and the greater part of the commercial 
supply is obtained from the distillation of coal for coke or gas ; of 
peat ; of bituminous shales ; of bones and refuse animal matter ; of 
putrid urine and excreta; of the residues from the fermentation of 
beet sugar molasses for alcohol; and the waste gases from blast 

Ammonia can be prepared synthetically from the nitrogen of the 
air in several ways, but these processes meet severe competition from 
the ammonia recovered as by-product. It was early proposed * to 
pass air over a mixture of barium oxide and carbon, at a white heat, 
and then to decompose the barium cyanide formed, by passing in 
steam after the temperature had been lowered to 450 C. : 

BaO +3C+2N+O= Ba(CN) 2 + CO 2 . 
Ba(CN) 2 + 3 H 2 O = BaO + 2 CO + 2 NH 3 . 

The process failed commercially because of the great consumption of 

The Frank and Caro process f is based upon the decomposition of 
calcium cyanamid (p. 267) by the action of superheated steam : 

CaCN 2 + 3 H 2 O = CaCO 3 + 2 NH 3 . 

Yields of 96 to 97 per cent are claimed, and the ammonia is very pure. 
The Haber process J depends on the realization of the reaction : 
N 2 + 3 H 2 = 2 NH 3 . Since this reaction takes place with decrease in 
volume and with a large evolution of heat (24 Cal.), it is driven to 
the right by increase in pressure, but tends to reverse at higher tem- 
peratures. While the equilibrium corresponds to high percentages of 
ammonia at ordinary temperatures, dissociation into the elements 
rises very rapidly with the temperature. Moreover, at low tempera- 
tures, the reaction rate is very slow. Haber has been able to find 
many catalyzers, of which he uses uranium powder and carbon, but 
even with these the rate is too slow for commercial work below 500 C. 

* Compt rendu, L, 1100. 

J. Soc. Chem. Ind., 1882, 364 ; 1883, 328. 

t Zeitschr. angew. Chem., 1903 (16), 536. J. Soc. Chem. Ind., 1903, 809 ; 1908, 
1093. Met. Chem. Eng., 1915 (13), 213. 

J Zeitschr. Elektrochem., 1910 (16) 244. J. Soc. Chem. Ind., 1910, 485, 1453. 



To get reasonable yields at this temperature, the pressure must be 
increased to 150 to 200 atmospheres. The reaction products, con- 
taining only a few per cent of ammonia, are cooled, the ammonia 
absorbed as a double compound with suitable salts, such as ammo- 
nium nitrate, and the gases returned to the reaction chamber. The 
ammonia is then driven off by heat, and the ammonium nitrate thus 

In Serpek's process * a mixture of calcined bauxite and coke 
is heated in nitrogen, or a producer gas containing 65 per cent of 
nitrogen, to 1600 or 2000 C., in a rotary furnace : - 

A1 2 O 3 + 3 C + N 2 = 2 A1N + 3 CO. 

The aluminum nitride, containing about 30 per cent nitrogen, is de- 
composed with water, according to the reaction 

2 A1N + 3 H 2 O= A1 2 O 3 + 2 NH 3 . 

The bauxite enters the upper one of two superimposed rotary kilns, 
and is heated by the combustion of the carbon monoxide issuing from 
the lower kiln ; the calcined alumina passes into the lower kiln, along 
with the carbon, and here the main reaction takes place. The alumi- 
num oxide produced is pure enough to meet market requirements. 

The chief source of ammonia is the " gas liquor " from the coke 
and gas manufacture (pp. 36, 314). The nitrogen in coal yields am- 
monia and cyanogen compounds by destructive distillation: in the 
" gas liquor " are found free ammonia, with ammonium carbonate, 
sulphide, and sulphydrate, which are volatile with steam, and sul- 
phate, thiosulphate, sulphite, sulphocyanide, and ferrocyanide, which 
are not volatile. Gas liquor is valued according to its percentage of 
ammonia as determined by distilling with caustic soda, absorbing the 
vapors in standard sulphuric acid, and titrating the excess acid. 

The gas liquor contains some tar, but on standing this settles, and 
the clear liquor is then distilled for the ammonia. In the simplest 
apparatus, the liquor is heated in one still until the volatile salts are 
expelled, and then is drawn into another still, where " milk of lime " 
is added, and heated until the fixed salts are decomposed and the 
ammonia driven off. The ammonia and volatile salts are absorbed 
in acid and the hydrogen sulphide and other foul-smelling gases evolved 
are led into the chimney, or decomposed in a Glaus kiln (p. 106). 

Generally, continuous stills constructed on the principle of the 

* U. S. Pat. 867,615, 888,044, 987,408, 996,032. Met. and Chem. Ind., 9 
(1913), 137. J. Soc. Chem. Ind., 1913, 1143. 



Coffey still, or some special apparatus, as that of Feldmann, or of 
Griineberg and Blum, are used to distil the gas liquor. Feldmann's 
apparatus (Fig. 72) is much used in this country : the gas liquor from 

the settling tank (F) 
passes into the econo- 
mizer (E), a long, cylin- 
drical shell, containing 
a number of narrow 
tubes, through which 
the liquor flows. In 
the vessel (D) is sul- 
phuric acid, to combine 
with the ammonia va- 
pors passing from the 
still by the pipe (G). 
The hydrogen sulphide 
and carbon dioxide lib- 
erated in (D) collect 
under the bell. The 
heat of the reaction 
between the acid and 
ammonia raises the 
temperature of these 
gases to a high degree, 
and they pass into the 
jacket or shell surrounding the tubes in the economizer, where they 
heat the liquor in the small tubes, so it arrives hot, at the top of the 
tower (AB) by the pipe (K). In the tower, the ammonia and its 
volatile salts are driven out by steam passing up through it. The 
liquor containing the fixed ammonia salts then passes to the lower 
part of (AB), where it is mixed with " milk of lime," while steam is 
blown in. The mixture then overflows through (M) into the smaller 
still (C) where all the ammonia set free by the lime is driven out by 
a steam jet from (S). This ammonia passes through (ON) into the 
first tower, mixes with the gas escaping from (AB), and is absorbed 
in (D). The waste liquor escapes through (P), and the sludge of 
calcium salts formed in (B) is drawn off at regular intervals through 
(R). The still may run for months without stopping. 

The Griineb erg-Blum apparatus is more complicated in details, 
but involves nearly the same principles as the above. All of these 
stills employ dephlegmation (p. 11). 

FIG. 72. 


Sometimes in distilling gas liquor, the vapors set free by the action 
of the lime are made to bubble through fresh liquor in a second vessel. 
Thus the volatile ammonia salts are expelled by the heat of these 
vapors, and pass with them to the absorption vats, while the gas 
liquor is drawn into the first vessel to be treated with lime. This 
method was used in the old apparatus of Griineberg and of A. Mallet. 

The ammonia gas set free in any of these stills is generally absorbed 
in sulphuric acid. If dilute acid (80 to 100 Tw.) is used, there is no 
separation of ammonium sulphate crystals in the saturator, and the 
liquor is easily clarified from tar and suspended impurities before 
evaporating to crystallize, and yields a light-colored product. With 
concentrated acid (140 Tw.) ammonium sulphate crystals separate 
in the saturator, and are " fished out." But they are often discolored. 
As the crystals are removed, fresh acid is introduced into the saturator. 
This is always covered with a hood, from which a pipe carries off the 
foul gases, consisting largely of hydrogen sulphide. These gases 
are often led to a Glaus kiln (p. 106) to recover the sulphur, and avoid 
contaminating the atmosphere. The ammonia gas is led into the 
saturator through a pipe perforated with small holes and submerged 
in the acid. 

Plants for distilling peat* in Mond producers (p. 43) to recover 
gas and ammonia are in operation abroad. The moist peat is treated 
with superheated steam and air at 350 to 500 C. ; the yield of am- 
monia varies from 40 to 130 kg. of ammonium sulphate per ton of 

Ammonia has been made in this country by distillation of waste 
animal matter from slaughter houses and tanneries.! The material is 
dried and put into an upright iron cylinder, provided with a manhole 
at the top and bottom, and having a large perforated pipe running up 
through the centre, about three-fourths the distance to the top. Chimney 
gases, forced by an air compressor through a superheater (a furnace 
containing coils of pipe heated to a bright red heat) into the perforated 
pipe, come into direct contact with the refuse matter. The volatile 
products pass out at the top of the retort into a hydraulic main, similar 
to that in a gas works. The tarry matter settles in the main and the 
gases pass through condensers. Both the condensed liquors and the 
gases pass into absorption tanks containing water ; the unabsorbed gases 
then go to a " scrubber " (p. 319) to remove the last of the ammonia, 
and are then burned under the retort. The liquor produced in the ab- 
sorbers and scrubber is distilled in an ammonia still. Much nitrogen 
remains in the coke in the retort. 

* Zeitschr. angew. Chem., 1906, 1574. J. Soc. Chem. Ind. 1908, 796 ; 1911, 744. 
t The process failed in practice and has been given up. 


A considerable amount of liquid ammonia is prepared for use in 
ice machines (p. 23). This is compressed into steel cylinders, usually 
containing about 100 pounds of the liquid. 

Ammonium sulphate, as found in commerce, has a light gray or 
yellowish color, or, if carefully made and washed after crystallizing, 
is nearly white. When prepared by direct saturation, the color may 
be brown or nearly black. Common acid made from pyrites yields 
a salt which is yellow in color, owing to the iron or arsenic present. 
The crystals should be washed, and dried in a lead-lined centrifugal 
machine. When sold in large quantities it is valued according 
to its content of ammonia or nitrogen. Good samples contain from 
23 to 25 per cent NH 3 . It is largely used as a source of nitrogen 
in making fertilizers, but for this purpose must be free from sulpho- 
cyanide, which is injurious to vegetation. When made by absorb- 
ing the gas in acid, little or no sulphocyanide is present, but by direct 
neutralization of the gas liquor the cyanide may separate with the 
sulphate. The salt is used as a source of other ammonium com- 
pounds, and to a slight extent in rendering fabrics, wood, and other 
tissues non-inflammable. By distilling with lime it yields a very 
pure ammonia gas, which may be absorbed directly in water for the 
" aqua ammonia " of trade ; or the gas may be passed through towers 
filled with charcoal, to remove any trace of pyridine or tar, before 
absorption. Any sulphuretted hydrogen may be removed by passing 
the gas over oxide of iron. 

Ammonium chloride is made by absorbing ammonia gas in dilute 
hydrochloric acid, or by neutralizing gas liquor with the acid directly 
and evaporating the solution. During the evaporation much of the 
tarry matter separates, and is skimmed off. Some nuisance may 
result from the gases escaping during the neutralizing. 

Another method is to mix a saturated solution of ammonium 
sulphate with a strong solution of salt or potassium chloride. On 
evaporating somewhat, monohydrated sodium sulphate (Na 2 SO 4 H 2 O) 
separates from the hot liquor, leaving the ammonium chloride in 
solution. On cooling, the ammonium chloride crystallizes : 

(NH4) 2 SO 4 + 2 NaCl = Na 2 SO 4 + 2 NI^Cl. 

The crystallized chloride is more or less discolored by tar, and 
is purified by sublimation in iron or earthenware pots or retorts. 
The ammonium chloride collects on the cover of the pot as a thick, 
fibrous cake, in which form it comes in trade under the name of sal- 
ammoniac. This generally contains iron as an impurity. It was 


formerly made by subliming the soot obtained by burning dried camel's 
dung, but is now nearly all made from gas liquor. The crystallized 
salt is often sold under the name of " muriate of ammonia," and is 
usually less pure than sal-ammoniac. Muriate of ammonia is much 
used in the arts for charging Leclanche electric batteries ; in the pro- 
cess of " galvanizing " iron ; in soldering liquors ; for making " rust 
cement " for pipe joints ; and in textile coloring. 

Ammonium carbonate as found in commerce is not a pure salt, but 
is a mixture of acid ammonium carbonate (NH 4 HCO 3 ) and a salt of 
carbamic acid (NH 2 -CO 2 NH 4 ). The commercial salt is made by 
heating a mixture of the sulphate and powdered calcium carbonate in 
iron retorts. The vapors are condensed in lead-lined chambers, and 
the impure product is generally sublimed in iron pots having lead caps. 
A little water is put into each pot along with the salt, this causing the 
sublimed product to be transparent instead of opaque white. The 
temperature of this second sublimation is not much above 70 C. 

Ammonium carbonate is transparent when fresh and pure, but on 
exposure to the air becomes covered with a white layer of bicarbon- 
ate, owing to the loss of ammonia. It is entirely volatile when 
heated, and from this fact is derived its old name of sal-volatile. It 
is used considerably in wool scouring, in certain baking powders, in 
medicine, and for the preparation of " smelling salts," and to some 
extent as an analytical reagent. 

Ammonium sylphocyanide (thiocyanate), p. 291. 


Acetic Acid, Vinegar, Ammonia, and Alum. John Gardner, F.I.C., 
F.C.S., London, 1885. (J. and A. Churchill.) 

Chemie des Steinkohlentheers. Dr. Gustav Schultz, 2te Auf., Vol. I, 
Braunschweig, 1886. (Vieweg und Sohn.) 

Das Ammoniak-Wasser. Albert Fehrmann, Braunschweig, 1887. (Vieweg.) 

Ammoniak und Ammoniak-Praeparate. Dr. R. Arnold, Berlin, 1889. 

Traitement des Eaux Ammoniacales. L. Weill-Goetz et F. Desor, Stras- 
bourg, 1889. (G. Fischbach.) 

Die technische Ausniitzung des atmospharischen Stickstoffes. E. Donath 
und K. Frenzel, Leipzig, 1907. 

Das Ammoniak und seine Verbindungen. J. Grossmann. Halle, a. S., 1908. 

Coal Tar and Ammonia. G. Lunge, 4th ed., London, 1909. (Gurney 
and Jackson.) 

Coal Gas Residuals. Frederick H. Wagner. New York, 1914. 
(McGraw-Hill Co.) 


Previous to the invention of the Leblanc Soda Process, the most 
important alkali was potassium carbonate, potash, which was 
nearly all derived from wood ashes. But with the development of 
the soda industry, the demand for potash was greatly diminished, 
and at the present time, soda has replaced it for all except a few 
special purposes. 

The chief sources of potassium salts are : 

Wood ashes. 

Beet-sugar molasses and residues. 

Wool scourings. (Suint.) 

Stassfurt salts. 

Land plants take up considerable quantities of potassium com- 
pounds from the soil. When the plants are burned, about 10 per 
cent of the weight of the ashes is potassium carbonate,* which may 
be obtained by lixiviation. Potash from wood ashes is now chiefly 
made in Russia, Sweden, and America, the woods most employed 
being elm, maple, and birch. Sometimes the stumps and small 
branches only are burned, the trunks being used for timber. The 
ashes are moistened slightly, put into tanks having false bottoms on 
which straw is spread, and then lixiviated with warm water. The 
lye so obtained is evaporated (sometimes by the waste heat from the 
burning wood) in iron pots until it solidifies on cooling. The dirty 
brown mass is then calcined in a reverberatory furnace until all the 
organic matter is destroyed. The product is known as potash or 
crude pearlash. It is white or gray in color, and contains about 
70 per cent K 2 CO 3 , with some sulphate and chloride and sodium salts. 
By redissolving the crude potash in water, settling and concentrat- 
ing the solution until the sulphates and chlorides separate as crystals, 
a concentrated and pure lye is obtained. When this is evaporated 
to dryness and the residue calcined, it yields a much purer product, 
known as " refined pearlash," and containing from 95 to 97 per cent 
of IQCOs. It is necessary that a low heat be employed in the cal- 
cination, since the charge fuses at a moderate temperature. 

Often, some quicklime is put in the bottom of the tanks before 
the ashes are introduced. On leaching, the solution of potassium 

* Those plants which contain much silica or phosphoric acid straw and 
grasses yield but little potash. 


salts reacts with the lime, forming insoluble calcium salts, and yield- 
ing more or less potassium hydroxide in the lye. The resulting prod- 
uct is then a mixture of potash and caustic potash. 

In the manufacture of beet sugar, a very impure molasses re- 
mains, containing among other things a large amount of soluble 
potassium salts. This molasses is now generally fermented, in 
which process the sugary substances are converted into alcohol, 
which is distilled off, leaving the mineral salts in the liquid resi- 
due, called mnasse or schlempe. If this is evaporated to dryness and 
the mass calcined, the organic potassium salts are decomposed, 
leaving in the cinder about 35 per cent potassium carbonate, and 
a large amount of chloride and sulphate, together with sodium 

If the mnasse be evaporated to dryness and the residue destruct- 
ively distilled in retorts, a distillate is obtained, containing organic 
compounds of which methyl alcohol, CH 3 OH, ammonia, and tri- 

/CH 3 
methylamine, N^-CH 3 , are valuable. The cinder in the retort con- 

X CH 3 

tains potassium salts, which are obtained in solution by lixiviation, 
and a considerable quantity of potash is thus recovered. Very often, 
however, the ash is used as a fertilizer, thus returning the potash 
to the soil. 

Wool scourings furnish some potash in countries where much 
wool is washed. Sheep's wool as it comes from the animal contains 
from 30 to 75 per cent of its weight of impurities, consisting of dirt, 
sand, dung, etc. ; wool grease or " yolk," a fat-like substance, made 
up of cholesterine and compounds of it with oleic, stearic, and palmitic 
acids ; and suint, which consists chiefly of potassium salts of 
oleic, stearic, and other organic acids, with small quantities of chlo- 
rides and sulphates and nitrogenous matter. The " suint " exudes 
from the animal in the perspiration, and is deposited on the wool 
by evaporation. It is soluble in cold water, and is thus removed 
in the scouring process. If these wash waters, containing wool 
grease and suint, are run into streams, pollution of the water results. 
Prevention of this nuisance, as well as the value of the potash, has 
necessitated disposal of the washings in some economical manner, 
and they are usually evaporated to dryness and calcined. If the 
calcination is done in closed retorts, a considerable quantity of am- 
monia is obtained. The cinder is lixiviated, and on evaporation, 
the solution yields, first, chlorides and sulphates of potassium and 


sodium, and finally a pure potash, which averages nearly 4 per cent 
of the weight of the raw wool scoured. For the recovery and treat- 
ment of wool grease, see pp. 369 and 500. This utilization of wool 
grease and suint is mainly practised in France, Belgium, and Ger- 
many, where it is done chiefly to prevent the pollution of the streams. 
Cheap fuel is essential to a successful working of the process. On 
a small scale it is not profitable, and the wash waters are often run 
on to the fields as fertilizer. 

For potassium carbonate from potassium chloride, see p. 162. 

Certain seaweeds, especially some varieties of brown algse, which 
grow in rather deep water, have the power of storing potassium salts 
in large amount. For many years the collection of seaweed in the 
kelp industry (p. 252) has been practised on the coast of Scotland and 
France, and from the ashes of these plants potassium chloride and 
sulphate, and iodine have been recovered. Recently the enormous 
beds of these plants (locally known as kelp) along the coast of Cali- 
fornia and in the Puget Sound region have attracted attention * as 
possible source for potash salts and iodine. The potassium salts 
may be extracted by diffusion processes, or by burning the plants at 
low temperatures and lixiviation of the ash. 

Reports * of large deposits of alunite (p 286) in Utah, Nevada, and Colo- 
rado have lately turned attention to the possibility of extracting potash 
from this basic alumino-potassium sulphate. By roasting, sulphur trioxide 
is evolved, the alumina rendered insoluble, and potassium sulphate may be 
lixiviated from the mass. The investigation of these deposits is not yet 

By far the most important source of potassium compounds at the 
present time is the great natural deposit of potassium salts found 
at Stassfurt and Leopoldshall, near Magdeburg, Germany. This 
consists of immense beds of various salts, which have been deposited 
from sea water. They were discovered in attempting to reach the 
underlying rock-salt, but because of the large proportion of potas- 
sium and magnesium chlorides, the material was at first thrown 
aside as worthless, the name applied to it, " abraumsalze," indi- 
cating the small value attached to it. But in 1861^ methods were 
devised by which potassium chloride and sulphate could be obtained 
cheaply from the Stassfurt salts, and since these furnish a valu- 
able source for nearly all other potassium salts, a rapid development 
of the industry followed. 

* Fertilizer Resources of the United States. Senate Document No. 190, 
62d Congress. Washington, 1912. 


Sea water contains about 3.5 per cent of solids, consisting of : 

Sodium chloride ......... 76.49 per cent * 

Magnesium chloride ........ 10.20 

Magnesium sulphate ........ 6.51 

Calcium sulphate ......... 3.97 

Potassium chloride ........ 1.98 

Magnesium bromide 1 85 " 

Calcium bicarbonate, etc. / 

By the evaporation of sea water under certain conditions, these 
salts, together with various double salts, formed by mutual inter- 
reactions, crystallize in the order of their relative insolubility. 

The Stassfurt deposit was undoubtedly formed by the evaporation 
of sea water, under peculiar conditions. The mode of formation has 
been studied by many investigators, to whose memoirs the reader is 
referred for full explanations, f The deposit is nearly 3000 feet thick, 
and about 16 different salts have been identified in the various strata. 
The more important salts and their composition, are given below : 

Gypsum ...... CaSO 4 2 H 2 O 

Anhydrite ...... CaSO 4 

Kamite ...... K 2 SO 4 , MgSO 4 , MgCl 2 - 6 H 2 O 

Carnallite ...... KC1, MgCl 2 6 H 2 O 

Kieserite ...... MgSO 4 -H 2 O 

Polyhalite ...... K 2 SO 4 , MgSO 4 , 2 CaSO 4 -2 ILO 

Rock-Salt ...... NaCl 

Sylvine ...... KC1 

Tachydrite ..... CaCl 2 , 2 MgCl 2 12 ILO 

Boracite ...... 2 (Mg 3 B 8 O 15 ) + MgCl 2 

Astrakanite ..... MgSO 4 , Na2SO 4 4 H 2 O 

Schoenite ...... K2SO 4 , MgSO 4 6 H 2 O 

The beds are not sharply defined layers of separate salts, the de- 
posit being generally regarded as containing four principal " regions." 

The-rock-s.alt or anhydrite region is the lowest of these. This 
consists of thin layers of very pure rock-salt, separated by narrow 
strata (one-fourth of an inch thick) of anhydrite. The anhydrite is 
separated from the salt mechanically, and the latter is then ground 
for use directly. This bed is nearly 2000 feet thick in places. 

The polyhalite region, about 200 feet thick, is above the rock- 
salt region. It is composed of 91 per cent of rock-salt, and 6j per 
cent of polyhalite, with smaller quantities of other salts. 

* Regnault (Thorpe's Dictionary of Applied Chemistry, Vol. IV, 340). 
t A very good account is given in Thorpe's Dictionary of Applied Chemistry, 
Vol. IV, pp. 340-341. Also see Pfeiffer's Handbuch der Kali-Industrie. 


The kieserite region, lying next above, is about 185 feet thick, 
and contains 65 per cent rock-salt, 17 per cent of kieserite, 13 per 
cent carnallite, and 5 per cent of other salts. 

The carnallite region lies nearest the surface, and is about 140 feet 
thick. This is the most important and contains ; 

Carnallite 55-60 per cent 

Rock-salt 20-25 per cent 

Kieserite 16 per cent 

Tachydritel 4 per cent 

Boracite j 

In parts of this region, changes have taken place through the ac- 
tion of water, by which considerable deposits of kainite and sylvine 
have been formed. The composition of raw carnallite is about as 
follows : j jj 

.. 15.7 per cent 

. . 21.3 

. . 21.5 

-, 0.3 

. --v. 13.0 " 

. . 00.0 

. . 26.2 

Insoluble 00.0 " , . . 2.0 

The crude carnallite is often colored a deep red by the presence 
of iron compounds. 

The present commercial supply of potassium chloride, and inci- 
dentally of other potassium compounds, is obtained from carnallite. 
The crude material is treated with the hot mother-liquor from a 
previous lot, in an iron kettle having a stirring apparatus and a false 
bottom. This mother-liquor contains about 20 per cent MgCl2, which 
prevents the solution of the rock-salt and kieserite, but does not 
hinder the dissolving of the carnallite. The action of the magnesium 
chloride solution is continued until the hot liquor reaches a density 
of 1.32 sp. gr., when it is drawn off from the sludge and allowed to 
cool slowly. At this density, the greater part of the potassium 
chloride crystallizes on cooling, leaving the magnesium chloride and 
some potassium chloride still in solution. This liquor is then further 
concentrated, until it contains about 30 per cent magnesium chloride. 
On cooling, crystals having the composition KC1, MgCl 2 6 H 2 O, 
artificial carnallite, separate, leaving only the excess of magnesium 
chloride in solution. The artificial carnallite is decomposed with 
water, and the potassium chloride crystallized out, leaving the mag- 

Potassium chloride 

. 16.2 per cent 

Magnesium chloride . 

. 24.3 

Sodium chloride . . 

. 18.7 

Calcium chloride . . 

. 0.2 

Magnesium sulphate . 

. 9.7 

Calcium sulphate . . 

. 2.1 




nesium chloride in solution ; a part of this liquor, diluted with the wash 
water from the sludge, is used to extract the next portion of raw carnal- 
lite. The potassium chloride is washed with a small portion of very 
cold water, to remove the common salt. 

The residue from the solution of the raw carnallite consists largely 
of kieserite mud (MgSO 4 H 2 O), which is insoluble in water; but on 
standing for some time in contact with water, it passes over into the 
soluble Epsom salts (MgSO 4 7 H 2 O). At an intermediate stage of 
the hydration, the mud solidifies in a manner similar to plaster of Paris 
when mixed with water. When this solidification is about to take 
place, the mud is moulded into blocks, which become very hard, and 
in which form it is shipped. But after some time they take up moisture 
from the air, and fall to a powder of Epsom salt. 

Glauber's salt is made at Stassfurt in the winter time as follows : 
Solutions of common salt and magnesium sulphate (e.g. from kieserite) 
when kept below C. will react together, thus : 

MgS0 4 + 2 NaCl = MgCl 2 + Na 2 SO 4 , 

and at the low temperature, the sodium sulphate crystallizes to form 
Na 2 SO 4 10 H 2 O. 

Kainite (K 2 SO 4 , MgSO 4 , MgCl 2 6 H 2 O) is extensively used in the 
crude state as a fertilizer. Some of it, however, is treated for potas- 
sium sulph'ate, by the method of H. Precht. When heated with 
water under pressures of four or five atmospheres, kainite decomposes 
into a double potassium-magnesium sulphate, magnesium chloride, 
and potassium chloride, thus :' 

3(K 2 SO 4 , MgSO 4 , MgCl 2 - 6 H 2 O) = 

2(K 2 SO 4 , 2 MgSO 4 - H 2 O) + 2 MgCl 2 + 2 KC1 + 16 H 2 O. 

The double potassium-magnesium sulphate separates in crystals, 
and is freed from chlorides by washing; during the washing, one 
molecule of the magnesium sulphate is also removed, and a salt of 
the composition, K 2 SO 4 , MgSO 4 , remains. This is dried and calcined 
and sold as double potassium-magnesium sulphate ; or it may be de- 
composed directly by treating with a solution of potassium chloride 
of 1.142 sp. gr. : - 

K 2 SO 4 , MgSO 4 + 2 KC1 = MgCl 2 + 2 K 2 SO 4 . 

The potassium sulphate is separated from the magnesium chlo- 
ride by crystallization. 


Potassium sulphate, made from kainite as above, or by the action 
of sulphuric acid on potassium chloride, is largely used as a fertilizer 
and for the manufacture of potassium carbonate. 

Potassium chloride, chiefly obtained from carnallite, is extensively 
used for preparing other potassium salts, especially the nitrate (p. 146), 
sulphate, and carbonate. 

Potassium carbonate or potash is made from potassium chloride 
by the Leblanc process, in the same way as soda-ash from salt. But 
the ammonia process cannot be employed, because the acid carbonate 
of potassium (KHCO 3 ) is soluble in ammoniacal solutions, and does 
not precipitate. 

Potassium carbonate is sold in trade under the name of potash 
or pearlash, and is used chiefly in the glass industry, for caustic potash 
and for chroma tes of potassium. A considerable quantity is bought 
by soap makers, and causticized, the solution being used for soft soaps 
(p. 373). 

Caustic Potash is made in the same way as caustic soda (p. 101). 
The mother-liquors from the black-ash lixiviation are decomposed 
directly with slaked lime. Caustic potash is much more deliques- 
cent than caustic soda, and is generally made where it is to be used. 

In soap making, it was formerly customary to saponify the fat 
with caustic potash, and then to add common salt. An interchange 
between the potassium and sodium took place, and a hard sodium 
soap resulted. But as soda is now cheaper, and yields a hard soap 
directly, potash soaps are only used for special purposes. 

Potassium nitrate (see p. 146). 

Potassium bichromate (K 2 Cr 2 O 7 ) is made by roasting chromite (a 
native oxide of chromium and iron) with potash, lixiviating the 
fused mass with water, and adding enough sulphuric acid to convert 
the neutral potassium chromate into bichromate. The reactions 
involved are as follows : 

Cr 2 O 3 + 3 O = 2 CrO 3 . 

CrO 3 + K 2 CO 3 = K 2 CrO 4 + CO 2 . 

2 K 2 CrO 4 + H 2 SO 4 = K 2 SO 4 + K 2 Cr 2 O 7 + H 2 O. 

The finely powdered chrome ore is mixed with lime and potash, 
and roasted at a bright red heat, with free access of air and frequent 
stirring. After several hours the chromic oxide is all oxidized to 
chromium trioxide (CrO 3 ), which combines with the lime and pot- 
ash to form neutral chromates of calcium and potassium. The mass 


is then treated with a hot solution of potassium sulphate, which 
forms potassium chromate from the calcium chromate. The solu- 
tion of neutral potassium chromate, when saturated, is drawn off 
and settled. It is then decomposed in lead-lined tanks, by the addi- 
tion of sulphuric acid. Since potassium bichromate is very much 
less soluble in cold solution than the neutral chromate, about three- 
fourths of the total amount of bichromate formed precipitates. The 
remaining liquor, containing potassium sulphate, is used to leach 
a new portion of cinder. The precipitated bichromate is recrystal- 
lized from water. 

The addition of lime to the furnace charge is necessary to prevent 
the fusion of the mass, and to keep it porous, so that the oxidation of 
the chrome is more complete. 

Potassium bichromate is much used as a source of other chromium 
compounds; as an oxidizing agent in dyeing and making coal-tar 
dyes ; as a mordant ; as -a bleaching agent for oils and fats ; and for 
the preparation of leather in the chrome tannage processes. 


Die Industrie von Stassfurt u. Leopoldshall. G. Krause. Coethen, 1877. 

Haudbuch der Kali Industrie. E. Pfeiffer, 1887. 

Die Salz Industrie von Stassfurt. Dr. Precht, Stassfurt, 1889. (R. 


Die Stassfurter Kali-Industrie. G. Lierke, Wien, 1891. (Hilschmann.) 
Die nprddeutsche Kaliindustrie. Precht-Ehrhardt, Stassfurt, 1906. 
Chemie und Industrie der Kalisalze. Erdmann, Berlin, 1907. 
Die deutsche Kaliindustrie. Dr. K. Kubierschky, Halle, a. S., 1907. 
Die Verwertung des Kalis. Kiersche, Halle, a. S., 1907. 
Fertilizer Resources of the United States. Senate Document No. 190, 

62d Congress. Washington, 1912. 
J. Soc. Chem. Ind., 1883, 146. C. N. Hake. 
Chemische Zeitung, 1890, Grief. 1891, Heyer. 
Dingler's polytechmsches Jour. Vol. 241. Precht. 


Growing plants abstract from the soil and air certain elements, as 
carbon, hydrogen, potassium, calcium, sulphur, phosphorus, and 
nitrogen, and apply them to their nourishment. To a less degree, 
silicon, iron, sodium, magnesium, and chlorine are taken up also. 
Natural weathering of the minerals in the soil usually provides enough 
of the elements needed by plants, but the supply of potassium, phos- 
phorus and nitrogen is insufficient for frequent repetitions of the 
same crops, and the soil becomes less productive * or barren. To 
supply this yearly drain on the soil, fertilizers are employed. The 
natural fertilizers, barn-yard manure, urine, and decomposing vege- 
table mould or muck, need little or no treatment before use, and will 
not be considered here. 

Artificial fertilizers are manurial substances prepared from mate- 
rials needing special treatment to render them fit for plant food. The 
chief requisites for a good artificial fertilizer are: It must contain 
at least one substance fit for plant food, and which is easily converted 
by rain or moisture into a form that plants can assimilate ; it must be 
dry and finely pulverized, for even distribution over the surface of the 
ground ; nothing injurious to plant life may be present. 

A complete fertilizer supplies the three essentials, potassium, 
nitrogen, and phosphorus. Often only one or two of these elements 
may be afforded, the fertilizer being intended for use with certain 
crops or on particular soils. . 

Potassium is generally returned to the soil in the form of sul- 
phate or carbonate (wood ashes), and occasionally as chloride. The 
preparation and use of these salts have already been considered and 
also the preparation of ground kainite (p. 161) for this purpose. 

Nitrogen is frequently supplied as ammonium salts (p. 154), or 
nitrates, particularly sodium nitrate (p. 145). But many substances 
used for fertilizers contain nitrogen in organic compounds, which 
decompose readily in the soil, setting free the nitrogen. 

Recently calcium cyanamid f (CaCN 2 ), made by treating calcium 

* Loss of fertility may be due to other causes then depletion of plant food : there 
is some evidence" that plants leave deleterious excretions in the soil, which for a 
time act toxically upon the same variety of plant. 

t Zeitschr. angew. Chem., 1903 (16), 536 ; 1910 (23), 2405. 
J. Soc. Chem. Ind., 1903, 809. 

Electrochem. Met. Ind., 1907, 77; 1908, 341 ; 1910, 539; 1915, 213. 



carbide at 1000 C. with nitrogen gas from liquid air (p. 267), has 
come into use as a nitrogenous fertilizer under the name " nitrolim" 

Phosphorus is nearly always applied to the soil in some form of 
calcium phosphate derived from mineral sources or from organic 

Fertilizers are largely made from the waste products of slaughter 
houses, such as blood, bits of waste meat and other refuse, bones, 
hoofs, horns, and hair. Tainted meat and animals which have died 
of disease are also sent to the rendering tanks.* Blood is dried at a 
moderate heat and crushed to powder between rolls. It contains 
about 10 per cent N, and is very uniform in composition. 

Raw bones contain fatty matter which is slow to decompose ; but 
if allowed to ferment in compost heaps with wood ashes and stable 
manure for a few months, they become more active and yield 3 to 
4 per cent nitrogen, as well as 20 to 25 per cent phosphate. As a rule 
bones are extracted with superheated steam to remove the fat and 
gelatine, and then ground to yield " bone meal," carrying about 27 
to 28 per cent of phosphoric acid. This is much more active as a 
fertilizer than the crushed raw bones. Steaming reduces the nitrogen 
to about 1 to 2 per cent, and makes the bone more easily decomposed 
in the soil. If treated with sulphuric acid, the nitrogen and phosphoric 
acid are rendered more available and the product is called " dis- 
solved bone." 

Bones are often subjected to destructive distillation in retorts, 
by which nearly all the nitrogen is driven out as ammonia, ammo- 
nium carbonate, pyridine, and other nitrogenous organic compounds, 
while the residue left in the retort, known as " bone-char " or " bone- 
black," contains calcium phosphates and other salts, mixed with 
carbon. This bone-char is extensively used as a decolorizing agent 
in the purification of sugar, glucose, oils, and other liquids ; when it 
can no longer be employed for this purpose (see p. 311), it is burned 
with free access of air to form " white-ash," which contains a high 
percentage of phosphorus. This bone-ash may be used directly as a 
fertilizer, but is usually treated with sulphuric acid to form " super- 
phosphate " (p. 169), which is more soluble than the tricalcium 
phosphate of the bone. 

A process for extracting the mineral phosphate from bones by 
digesting with hydrochloric acid has been practised to some extent. 

* " Rendering " consists in extracting all the fats, oils, and gelatinous matter 
from the carcasses by treating with benzine or steam under pressure. The fat 
extracted is used for soap stock. The dried residue, called " tankage," is ground 
fine for fertilizer, and furnishes both nitrogen and phosphoric acid. 


The solution of phosphoric acid thus obtained is neutralized with 
milk of lime, by which the calcium phosphate is precipitated, chiefly 
as dicalcium phosphate (Ca2H2P2Og). This is sometimes sold as 
" precipitated phosphate," but the method is more commonly applied 
to low grades of mineral phosphates (p. 171) than to bones. 

Garbage containing fatty matter is now collected in many cities 
and subjected to a rendering process. It is put into steel digesters 
and subjected to the action of steam at 50 pounds pressure for eight 
or ten hours, when the mass is reduced to a soft pulp which is put into 
presses and the oily matter pressed out. The press-cake is broken 
up and dried in revolving steam-heated drums, after which it is 
powdered, sifted, and used for " filler " in fertilizers, under the name 
" tankage." * It contains nitrogen, phosphoric acid, and a little 
potash. On cooling, the oily matter forms a soft grease, which is used 
for soap and candle stock. The water which is pressed out of the 
tankage with the grease contains a large amount of ammonium salts 
and some potash ; it is evaporated to dryness and the residue mixed 
with the tankage, thus increasing the nitrogen and potash in the 

Other nitrogenous wastes from various industries leather scrap, 
wool waste, and dust from shoddy and felt mills are used to some 
extent; but these, though very rich in nitrogen, are very slow in 
decomposing, and are so light when powdered that they are easily 
blown away. 

The press-cakes from various oil industries (e.g. the manufacture 
of cotton-seed, rape, and castor oils) are often ground for fertilizer. 
Sometimes the cake is burned for fuel and the ashes used for fertiliz- 
ing, but in this case the nitrogen is lost, only the potassium and 
phosphorus being returned to the soil. In the manufacture of fish 
oils there is a considerable amount of residue from which the oil has 
been pressed (p. 363). This is known as " fish scrap," and consists 
of the scales, bones, fins, and meat of the fish. It contains about 7 
per cent of nitrogen and nearly 16 per cent of phosphorus pentoxide 
(P2O 5 ). Dried (usually by exposure to the sun) and crushed to a 
rather coarse powder, it is a valuable fertilizer, decaying rapidly in 
the soil. 

Peruvian guano, formerly of great importance as a fertilizer, but 

the beds now nearly exhausted, consists of dried excrement, feathers, 

and carcasses of sea fowl, and is rich in nitrogen and phosphoric 

acid. It is found in certain islands near the coast of Peru and Chili, 

* This term is also applied to the dried residues of various rendering processes. 


and also on the mainland at the base of the Andes, near the sodium 
nitrate beds (p. 145). The region is dry and hot, and the guano 
has been preserved with a high percentage of nitrogen, largely as 
uric acid and its. salts. It needs no preliminary treatment before 
spreading on the soil. Fresh guano, from various islands in the 
South Pacific, is damp, and contains much ammonium carbonate; 
this must be " fixed " by mixing with sulphuric acid, to prevent loss 
of the nitrogen. 

Fossil guanos, consisting of fossil excrement and remains of birds 
and reptiles, are found in the West Indies, Bolivia, Chili, and the 
South Pacific islands. Since more or less rain falls in these climates, 
the soluble ammonium salts and nitrates have been washed out, 
leaving only the calcium phosphate. Some of these guanos have 
entered into combination with the rocks on which they were de- 
posited, thus altering their original character considerably ; e.g. some 
of them contain a large amount of calcium sulphate. Fossil guanos 
are prepared in the same way as phosphate rock (see below). 

Phosphoric acid is chiefly supplied by phosphate rocks, such as 
apatite, or phosphorite, found in large deposits in Belgium, Germany, 
France, Spain, Algiers, Canada, South Carolina, Florida, Tennessee. 
Arkansas, Montana, Wyoming, Idaho, Utah, the West Indies, and 
certain islands in the Pacific and Indian oceans. At present the 
United States deposits are the most important. 

Apatite [3 Ca 3 P 2 O 8 + CaF 2 (CaCl 2 )] is a crystalline mineral, 
occurring in large deposits in Canada and Spain. The former are 
very extensive, and are found in Ontario, between the St. Lawrence 
and Ottawa rivers, and in Quebec Province, along the Gatineau 
and du Lievre rivers. The mineral sometimes occurs in veins and 
pockets (bonanzas) of nearly pure, massive apatite; and in other 
cases as distinct, hexagonal crystals or nodules, disseminated in 
calcite or pyroxene. The material is sold on a guarantee of 75 or 
80 per cent of calcium phosphate, and to secure this degree of purity, 
" cobbing " * and hand-picking must be employed. The ore being 
exceedingly brittle and the gangue rock hard, there is much loss in 
the " fines," from which it is not profitable to separate the phosphate 

Apatite varies in character from a moderately hard rock to a soft 

and friable mass, called " sugar." The color varies much, but is 

generally blue-green or red-brown. The tricalcium phosphate being 

quite insoluble, the mineral must be treated with sulphuric acid to 

* Breaking the large lumps with hammers by hand. 


form " superphosphate." But since more or less calcium fluoride 
and chloride are present, considerable acid is uselessly consumed, and 
a special condensing apparatus is necessary to retain the vapors of 
hydrofluoric and hydrochloric acids set free, or a nuisance is created. 
Apatite also requires a rather strong acid (1.78 sp. gr.) for its decom- 
position, while the calcite and other minerals connected with it, being 
acted upon, cause considerable loss of acid. 

These objections do not apply to the phosphorites of the United 
States and Europe, and the cost of mining is not so great. As a re- 
sult the use of apatite for fertilizer making has ceased. 

Phosphorites are amorphous rocks of varying composition, but all 
containing a large percentage of tricalcium phosphate, and some- 
times iron and aluminum phosphates. The mode of formation of 
these rocks has been a much-disputed question, but they are now 
generally regarded as of organic, and probably animal origin. The 
beds are filled with fossil remains of land and marine animals and 
fishes. A nodular variety found in England was erroneously sup- 
posed to be fossil reptilian excrement, and was called " Coprolites." 

Some phosphorites are compact and hard to grind, as is the 
Spanish variety, but the American rock is softer and porous. In 
the United States there are two varieties, " land rock " and " river 

Land rock occurs in beds averaging from 10 to 12 inches in thick- 
ness, and from 2 to 40 feet below the surface of the ground. These 
beds are sometimes composed of loose pebbles or gravel, but frequently 
these have been compacted into solid layers having a laminated struc- 
ture; or they may form great boulders or conglomerate masses. 
The beds are often continuous over a large area, but " pockets " or 
isolated beds are frequently found. Good rock will average from 
75 to 80 per cent of tricalcium phosphate (CazPzOs). In some cases 
the land rock is hard, dense, and nearly pure (hard phosphate), while 
in others it is soft, resembling clay in its consistency, and usually 
containing rather a large proportion of iron and aluminum. 

Land rock is mined by stripping off the overlying earth, and 
digging out the phosphate rock with pick and shovel. It has been 
found practical to use steam shovels and dredges for " soft phosphate " 
and " pebble " deposits. In compact rock, blasting is necessary. 
The work is done in open pits, tunnelling not having proved success- 
ful. The depth of overburden which may be profitably removed 
depends upon the thickness and purity of the deposit, but about 
20 feet is the limit, except in the case of very thick beds of high-grade 


ore. For ordinary rock, the limit is about 10 or 12 feet. In a few 
cases hydraulic mining has been employed to wash away the over- 

After mining, the rock is put through a " breaker," and reduced 
to lumps about 4 inches in diameter. These go to the " washer," 
which consists of a long, semicircular trough, set at a slight incline, 
in which there is a revolving shaft, carrying teeth or blades about 
9 inches long, and arranged around it in the form of a spiral screw, 
having a pitch of about 1 in 6. The trough is set in a tank of water, 
or a large stream of water enters at the upper end. The lumps of 
rock are fed into the trough at the lower end, and being caught by 
the teeth, are forced along and up the trough, against the water. The 
rubbing against each other and the action of the water wash away 
the sand and clay, and at the upper end the clean rock falls on screens, 
which separate the several sizes of lumps. It is usually dried by pil- 
ing it on racks of cord wood, which are then fired and allowed to burn 
out ; or it may be piled over cast-iron pipes having numerous aper- 
tures through which hot air from a furnace is forced. The rock is 
then shipped to the makers of " superphosphate." 

River rock is dredged or dug from the beds of rivers and streams, 
especially Peace River and its tributaries in Florida, and from the 
streams near Charleston and Beaufort, S.C. When the deposit 
is in the form of loose nodules and gravel, steam dredges or cen- 
trifugal pumps are used to raise it; but when it is compact rock, 
special forms of grips and dredges are necessary. In most cases, 
river mining is not carried on in water more than 30 feet deep. 

River rock is very similar in composition to land rock, but is 
darker in color, even black, and contains more animal remains and 
fossils. It is preferred by foreign superphosphate makers and is 
generally shipped abroad. 

" Superphosphate " is the name given to a soluble phosphate, pre- 
pared by treating insoluble rock or bone * phosphate, with sulphuric 
acid. By the action of the acid, the insoluble tricalcium phosphate 
is converted into monocalcium phosphate (CaH 4 P 2 O 8 ), while in 
many cases some free phosphoric acid is also formed. 

The reactions involved are as follows : 

1) Ca 3 P 2 O 8 + 2 H 2 SO 4 + 4 H 2 O = CaH 4 P 2 O 8 + 2 (CaSO 4 2 H 2 O). 

2) Ca 3 P 2 O 8 + 3 H 2 S0 4 + 6 H 2 O = 2 H 3 PO 4 + 3 (CaSO 4 2 H 2 O). 

3) Ca 3 P 2 O 8 + H 2 S0 4 + 4 H 2 = (Ca 2 H 2 P 2 O 8 - 2 H 2 O) + (CaSO 4 2 H 2 O). 

* Superphosphate made from bones contains some nitrogen as well as phosphoric 


Reactions (1) and (2) are the ones desired in fertilizer making, but 
if too little acid is used, reaction (3) takes place to a greater or less 
extent, forming dicalcium phosphate, which is also insoluble. If 
too much acid is used, reaction (2) takes place to an undesirable ex- 
tent, and the product contains an excess of free phosphoric acid, 
which attracts moisture from the air, making the fertilizer moist 
and lumpy. A small excess of acid over the theoretical quantity 
needed is generally used to prevent "reversion" (below) as far as 
possible. The proper regulation of the amount of acid requires 
great care, and must be controlled by analysis of the material. The 
acid employed is "chamber acid " of 1.54 to 1.60 sp. gr. Concen- 
trated acid is not used, because water is necessary in order that a 
hydrated calcium sulphate may be formed. The formation of the 
gypsum (CaSO4 2 H 2 O) aids in the subsequent drying of the product. 

Hydrochloric acid is unsuitable for fertilizer making, because of 
its expense, and the formation of calcium chloride in the product. 
The raw phosphate should be as free as possible from impurities, 
such as carbonates, iron oxide, and alumina. About 3 per cent of 
Fe 2 O 3 + A1 2 O 3 is the limit now allowed. 

The phosphate rock, ground in Griffin, Kent, or ball mills (p. 188) 
to pass a 60- or 80-mesh sieve, is put, with the required amount of 
acid, into a cast-iron mixer, provided with a stirring device. The 
mixing is complete in two to five minutes, when the slimy mass is at 
once run into a brick-lined " pit " or " den," where the reactions 
take place. The temperature rises to 100 or 110 C., and much fume 
(HF, SiF4, CO 2 ) escapes. As the reactions progress, the charge stiffens 
and finally solidifies into a porous dry mass. Successive charges from 
the mixer are dumped into the " den " until it is filled ; then the whole 
is left quiet for some days or weeks, for the reactions to end. The 
product is then dug out of the pit, pulverized in a disintegrator, where 
nitrogenous or potash materials may be added if desired, and packed 
in bags for market. 

If the phosphate rock contains much iron or aluminum oxide, or 
if the decomposition by acid has been incomplete, a series of sec- 
ondary reactions ensues when the superphosphate is stored. By 
these, a part or all of the monocalcium phosphate (CaH 4 P 2 O 8 ) and 
the free phosphoric acid may be converted into the insoluble di- 
calcium phosphate, or into insoluble phosphates of iron or aluminum. 
This constitutes " reversion," and the insoluble calcium or iron 
phosphates so formed are called " reverted phosphate." Since 
fertilizer is valued according to its percentage of soluble phosphate, 


reversion is a serious matter for manufacturer and buyer. Reverted 
phosphate is recognized as having value for fertilizer purposes, but 
less than superphosphate. 

When due to incomplete decomposition of the rock, reversion 
takes place according to the following reaction : 

CaH 4 P 2 O 8 + Ca 3 P 2 O 8 = 2 Ca 2 H 2 P 2 O 8 . 

When the rock contains iron or alumina, the temperature of the 
reaction in the pit is kept as low as possible, to prevent combina- 
tion between these oxides and the free phosphoric acid formed. It 
is customary in this case to remove the superphosphate from the pit 
as soon as it solidifies, and to cool it by exposure to the air. 

A " double superphosphate " is also made, in Europe, and contains 
more soluble phosphoric acid than the ordinary superphosphate. 
A quantity of bones or phosphate rock is 'decomposed with sufficient 
dilute sulphuric acid to set free all the phosphoric acid and precipi- 
tate all the calcium as hydrated calcium sulphate. The precipitate 
is removed by the filter press (p. 15), and the clear solution of phos- 
phoric acid is concentrated by surface heating in lead pans, to a den- 
sity of 45 Be., at which strength the solution contains nearly 45 per 
cent P2O 5 . During concentration, the iron and aluminum phos- 
phates separate and are removed. The strong solution of phosphoric 
acid is then treated with ground phosphate rock, in proper quantity 
to form monocalcium phosphate, which is dried and disintegrated. 
The reactions are as follows : 

1) Ca 3 P 2 O 8 + 3 H 2 SO 4 + 6 H 2 O = 3(CaSO 4 2 H 2 O) + 2 

2) Ca 3 P 2 O 8 + 4 HaP04= 3 

By this process, a very concentrated fertilizer, containing no 
gypsum or other sulphate, is obtained. Moreover, a low-grade phos- 
phate rock, which would not furnish a strong fertilizer with sulphuric 
acid, can be used for making the phosphoric acid. 

Phosphate rock is also used directly for fertilizer, without other 
preparation than fine grinding. But tricalcium phosphate, being in- 
soluble, is only slowly assimilated by plants, and its action is not very 
marked. Several years are necessary for its complete decomposition. 

Phosphatic slag is now used to a considerable extent as a fertil- 
izer, especially in Europe. In the process of making Bessemer steel 
by the Thomas and Gilchrist method, pig iron from ores containing 
phosphorus is .treated with an excess of lime in a Bessemer con- 
verter, lined with lime, while a blast of air is forced into the liquid 


mass. At the high temperature of the melted iron, the phosphorus 
is oxidized to pentoxide, which combines with the lime. The silica, 
alumnia, lime, and magnesia unite to form a slag, into which the 
calcium phosphate produced also goes. By proper regulation of the 
charge, a slag containing about 17 per cent of pentoxide (PzO^) is 
obtained. The phosphate in the slag is supposed to be a tetracalcic 
phosphate (Ca 4 P2O 9 ), which is insoluble in water, but is much less 
stable than tricalcium phosphate. When exposed to the weather in 
the soil, it decomposes, though somewhat slowly, and the phos- 
phorus passes into a form which plants can assimilate. In order 
that this decomposition may take place, the slag must be ground 
very fine, so that 90 per cent of it will pass through a sieve with 
100 meshes to the linear inch. The grinding is best done in a ball 
mill (p. 187). 

Slag fertilizer needs no further treatment than very fine grind- 
ing, but it is slow in decomposing, and its full effect is not obtained 
for two or three years. It decomposes more rapidly than ground 
phosphate rock, however, and is cheap. 

There has been considerable controversy among agricultural 
chemists as to the relative value of soluble and insoluble phosphates. 
Some hold that the soluble phosphate is at once converted into the 
insoluble form when it comes into contact with the lime, alumina, 
and iron in the soil ; and that this insoluble phosphate is dissolved 
or absorbed by the sap in the plant roots, the sap presumably having 
an acid nature. Other chemists claim that only the soluble phos- 
phate, as such, can be taken up by the plant. It appears from ob- 
served facts that both soluble and insoluble phosphates are taken 
up by the plant, but the nature of the soil is an important factor. 
On a soil poor in lime, and containing some organic matter, insoluble 
phosphates produce their best results ; but if the soil contains much 
lime, then the superphosphate appears to have the advantage. 

The soluble character of the superphosphate permits its dif- 
fusion through the soil by rain, so that it is brought immediately 
to the roots of the plants. But the insoluble phosphate must be 
turned under the soil, and the roots grow to it ; then, too, when not 
finely ground, it possesses but little value, owing to the slow decom- 
position ; but when in a very fine powder, it is taken up in some way 
by the roots of the plant with fair rapidity. 

The manufacture and sale of artificial fertilizers are, to a certain 
extent, under legal restriction in nearly all the states. To prevent 
fraud, manufacturers are required to take out a license, and to sub- 


mit samples for analysis by state chemists; frequently a guarantee 
of the stated composition is required. 

The methods of analyses of fertilizers are set forth in detail in the 
bulletins of the several state agricultural experiment stations and of the 
United States Department of Agriculture.* In general the matter de- 
termined by the analysis may be summed up as : 

(a) Water, both hygroscopic and combined. 

(6) Total phosphoric acid. 

(c) Soluble phosphoric acid. 

(d) Reverted phosphoric acid. 

(e) Total nitrogen. 
(/) Potash. 

Another substance frequently sold as fertilizer is pulverized 
gypsum (CaSO 4 2 H 2 O), which, when crushed to a fine powder, is 
brought into commerce under the name of " plaster." As a fertil- 
izer it is of little value, except in soils poor in lime or those contain- 
ing "black alkali" (sodium carbonate). But it is also claimed to 
have a beneficial action in retaining nitrogen in the soil. The cal- 
cium sulphate is supposed to be decomposed by the ammonia and 
carbonic acid from the air and rain, forming ammonium sulphate 
and calcium carbonate. Ammonium sulphate furnishes nitrogen in 
a form which plants can assimilate. 

Much attention has been devoted, especially in Germany, to 
methods of recovering fertilizing material from the sewage of cities. 
But when closets are flushed with water, the effluent is generally too 
dilute to be worth recovering. It is, however, used to some extent, 
in irrigating lands, generally those owned by the municipality. Sew- 
age is often precipitated with lime or other substance, but this gen- 
erally renders the sludge useless for fertilizing. The contents of dry 
vaults or cesspools are collected at regular intervals and used for 

But sewage treatment of any kind is usually practised to pre- 
vent pollution and unsanitary conditions in the streams and water 
supplies, rather than for the utilization of the fertilizing materials 
to be obtained. 


Lehrbuch der Diingerfabrication. Paul Wagner. Braunschweig, 1877. 
Report of the Commissioner of Agriculture of South Carolina, for the year 

1880. Chas. U. Shepard. 

Bulletin de la Societe Chimique, 1884, 219. E. Dreyfus. 
Die kiinstlichen Dungermittel. Dr. S. Pick, Leipzig, 1887. (Hartleben.) 

* Bull. No. 28, U. S. Dept. Agriculture ; Division of Chemistry. 


The Nature and Origin of Deposits of Phosphate of Lime. R. A. F. Pen- 
rose, Bull. 46, U. S. Geological Survey, Washington, 1888. 

A Treatise on Manures. A. B. Griffiths, London, 1889. (Bell and Sons.) 

The Phosphates of America. Francis Wyatt, New York, 1891. 

Florida, South Carolina, and Canadian Phosphates. C. C. Hoyer Millar, 
London, 1892. (Fischer and Co.) 

Les Phosphates de Chaux naturels. Paul Hubert, Paris, 1893. 

The Phosphate Industry of the United States. Carroll D. Wright, 
Washington, 1893. (Sixth Special Report of the U. S. Commissioner 
of Labor.) 

J. Amer. Chem. Soc., 1893, 321. C. U. Shepard. 1895, 47. W. E. 

J. Soc. Chem. Ind., 1888, 79. W. T. MacAdam. 1894, 842. Kalmann 
and Meissels. 

Agricultural Analysis. H. W. Wiley, 2d ed., Vol. II, Fertilizers, 1908. 

Die Superphosphatfabrikation. Dr. Ritter von Grueber, Halle, 1907. 

The Manufacture of Chemical Manures. J. Fritsch, New York, 1911. 

Fertilizer Resources of the United States. Senate Document No. 190, 
62d Congress. Washington, 1912. 


Good lime is nearly pure calcium oxide; it is one of the most 
important substances used in chemical industry, and is prepared in 
enormous quantities by calcining calcium carbonate (limestone, 
chalk, or the shells of mollusks) at a bright red heat. If the carbon- 
ate used contains much silica, iron, alumina, or other impurity, the 
lime does not slake freely with water, and is said to be " poor " or 
" lean " ; with but small quantities of these impurities present, a fair 
lime is produced, when properly burned. Such impure carbonates 
are difficult to burn, as slight overheating causes semi-fusion of the 
lumps, and the lime combines with water slowly and incompletely, 
and is said to be " burned to death." A pure lime, which combines 
readily with water to form a fine white powder, free from grit, and 
which makes a smooth stiff paste with an excess of water, is called a 
"fat "lime. 

Calcium carbonate begins to decompose below a red heat into 
calcium oxide and carbon dioxide, but the decomposition is not 
complete until a bright red heat (800 or 900 C.) is reached. The 
temperature should not rise much above 1000 to 1200 C., as there 
is danger of overheating the lime. It is essential that the gases es- 
cape freely from the kiln, the draught usually being sufficient to 
remove them as they form. This escape may be accelerated by blowing 
steam or air into the kiln during 
the burning, or even by wetting 
the carbonate as it is introduced. 
If the gases are retained, they 
cause pressure in the kiln and 
thus hinder the decomposition ; 
and on cooling, the carbonic acid 
recombines with the lime. 

Limekilns are of two classes, 
periodic and continuous (see p. 
22). In this country, long-flame, 
periodic kilns are sometimes used, 
though they are uneconomical of ' 
fuel and time : but they are em- 
ployed because of simplicity and cheapness of building. They are made 
of brick or large stone blocks, an arch (A, Fig. 73) being turned two 




or three feet from the ground, with numerous openings left for the flames 
to pass into the kiln. The fire burns under the arch, on top of which 
the limestone is piled, the lumps varying from the size of a cocoanut 
just above the arch, to that of a goose egg at the top of the kiln. 
In starting the kiln the temperature is slowly raised during six or eight 
hours, to prevent the limestone arch from crumbling ; then a full red 
heat is held for two days or more, when the fire is allowed to burn out 
and the kiln cools. During the time of cooling, discharging, and re- 
charging, the kiln stands idle, thus losing much time. Moreover, a 
large amount of fuel is required to heat the walls of the kiln after 
each recharging. 

Continuous kilns are preferred where fuel is expensive, and where 
a large, regular output is desired. They are tall furnaces (shaft 
kilns), built of brick or of iron plates, and 
usually from 40 to 50 feet high, by 6 to 10 
feet diameter. The limestone is fed in at the 
top and the lime taken out at the bottom 
without interrupting the process. Vertical 
shaft kilns for mixed feed of fuel and lime- 
stone, or separate combustion of the fuel, are 
frequently used. The mixed feed kiln is 
cheaper to build and operate, but yields a 
product somewhat unevenly burned and dis- 
colored. The separate combustion type is 
commonly used at the larger plants, and while 
of lower fuel efficiency, it yields clean, high- 
grade lime. Such a kiln (Fig. 74)* consists 
of a steel shell, lined with fire-brick (B) forming 
a shaft 6| feet inside diameter and 48 feet 
high. There are four furnaces (A), the flames 
from which enter the shaft at (D) and pass 
up through the limestone in the shaft. From the cooling cone (C), the 
lime is dropped at intervals through the draw-gate into the car 

The separate combustion type of kiln works best with a long- 
flame combustible, such as wood, soft coal, oil, or gas. Otherwise the 
only heat supply for the reaction is from the hot gases from the grate, 
from which the heat absorption is slow. A volatile combustible 
may be burned in immediate contact with the charge, even within the 
pores of the lime itself, evolving its heat exactly where needed. A 
* Courtesy of Steacy-Schmidt Manufacturing Company, of York, Pa. 

FIG. 74. 



recent improvement* in kiln construction (Doherty-Eldred Kiln, 
Fig. 75), f is the use of induced draft by a fan drawing the gases from 
the top of the kiln, diverting a part of them 
into the air supply under the grate. There the 
carbon dioxide is reduced to carbon monoxide 
by the coal, and is later reoxidized by excess 
air in the kiln itself. This secures a long- 
flame action of the combustion, producing 
uniform heating, with less destruction of the 
furnace lining and less overburned lime. 

The best fuel for lime burning is wood, 
since it gives a long flame of only moderate 
heat intensity; but owing to its cost, other 
fuels, as coal, coke, oil, and natural and pro- 
ducer gas, are used. Oil arid gas firing yields 
very clean lime, burned at a constant tem- 
perature. To calcine one kilo of calcium 
carbonate requires 425 Cal. : thus taking the 
thermal value of carbon as 8140 Cal., it 
appears that each 100 kilos of limestone, or each 56 kilos lime pro- 
425 x 100 

FIG. 75. 

duced, will require 

= 5.2 kilos of carbon as fuel. This 


takes no account of the fuel needed to heat the limestone to the 
temperature of dissociation, nor of losses by radiation and conduc- 
tion in the burning. In practice, for each 100 kilos of lime produced, 
the kilns consume from 16 to 20 kilos of good coal, and with periodic 
kilns this may be nearly doubled. 

Excepting the limekilns in the ammonia-soda works, no attempt, 
as a rule, is made to save the carbonic acid gas which escapes from the 
top of the kiln. But in Europe the gas is often collected and used for 
technical purposes. 

Freshly burned lime is called " caustic " lime, or " quicklime," 
because of its corrosive action on organic matter. When pure, it 
is white and amorphous, but iron gives it a yellow tint. The crys- 
talline limestones and pure marble yield the best lime. Owing to 
the loss of water, organic matter, and carbon dioxide during the 
burning, there is great reduction in the weight of the charge, but only 
a slight decrease in its volume. As a rule, 100 pounds of good lime- 

* This is an application of the Eldred combustion process. Eng. Pat. 17,197 
of 1901. Jour. Soc. Chem. Ind., 1902, 696. 

t Catalogue of the Improved Equipment Co., of New York. 



stone yield about 56 pounds of lime, but the shrinkage in bulk is not 
over 10 to 20 per cent of the original volume of the stone. The 
hardness decreases and the lime is much more porous than the lime- 
stone, absorbing considerable water before slaking. Lime has great 
affinity for water, and when wet the lumps expand and fall to powder 
of calcium hydroxide (slaked lime), with the evolution of much heat, 
especially in the case of " fat lime." When exposed to the air, lime 
absorbs carbon dioxide and moisture, and soon falls to a powder called 
" air-slaked lime," consisting of a mixture of calcium carbonate and 
hydroxide. Lime for mortar and other purposes is generally slaked 
immediately before use. Pure lime is infusible at the temperature 
of the oxy hydrogen flame (hence its use in the "calcium light"), 
but if silica, iron, alumina, or other impurity is present, the lime 
combines with it to form a slag or glass. Lime is a powerful base and 
combines with acids to form calcium salts. 

Magnesium carbonate is a common associate of limestone, but 
if there is not more than 5 per cent of magnesia present, the 
burned product is called a high-calcium lime ; if 30 per cent or more 
magnesia is present, the lime is " high-magnesium " lime. Dolomite 
(CaCO 3 MgCO 3 ) and high-magnesian limestones generally appear 
to dissociate at lower temperatures * than do the pure limestones. 
Magriesian limes slake slowly and with less heat evolution than pure 
lime, and they make a stiffer mortar paste. 

Certain siliceous or argillaceous limestones yield a product, upon 
burning, which slakes and falls to powder when treated with water ; 
but after a time the pulverized mass becomes hard and rocklike 
through further action of the water upon the calcium silicates or 
aluminates, produced during the burning. Such lime is called " hy- 
draulic lime " and forms an intermediate link between the common 
limes and the true cements. The calcination is conducted at medium 
temperature but high enough to cause combination between some of 
the calcium oxide with nearly all of the silica and alumina, and still 
leave enough free lime to slake the clinkered material. Limestone 
suitable for hydraulic lime usually contains from 70 to 80 per cent of 
calcium carbonate, with 12 to 17 per cent of silica, and less than 3 per 
cent of alumina and iron together. Hydraulic limes are chiefly em- 
ployed in mortar and cement mixtures. 

Since much heat is liberated in the slaking of lime, the storage 
and shipment are attended with some danger. If water comes into 
contact with the lime in presence of combustible material, fire is very 
apt to ensue. 

* Probably about 600 to 700 C. (Eckel). 


Hydrated lime, slaked with just the proper quantity of water to 
yield a dry fine powder, is prepared commercially. The crushed 
lime is slaked in mechanical hydrators, sometimes of a form similar 
to the chlorine absorber (Fig. 63). The water and lime enter the top 
cylinder in measured amounts, and the hydration proceeds to com- 
pletion. Batch hydrators, consisting of a revolving pan with ploughs 
to mix the charge of lime and water, are also used. After hydration 
the slaked lime is screened and ground to fine powder, and then packed 
for market. Hydrated lime does not deteriorate so rapidly in stor- 
age and the fire risk is greatly reduced. It can be mixed with cement 
and other materials, where quicklime is not suitable. 

The following are a few of the important uses of lime in the arts : 
in mortar and cement mixing ; in bleaching powder ; in the alkali 
manufacture ; for purifying illuminating gas ; in the preparation and 
purification of many chemicals, such as acetic, citric, oxalic, and tar- 
taric acids, caustic soda and potash, etc. ; for purifying sugar solutions ; 
in bleaching cotton ; in tanning ; in glass making ; in metallurgical 
operations; for disinfecting, etc. 

Mortar is an aqueous pasty mixture of slaked lime, sand, and 
other materials, which dries without excessive shrinkage and be- 
comes hard on exposure to the air, owing to absorption of carbon 
dioxide and formation of carbonate of lime. It will not harden 
while it remains very wet, and this is one of the chief differences be- 
tween mortar and cement. The hardening of the two substances is 
due, in part at least, to different causes. 

If a paste of freshly slaked lime is allowed to dry by exposure to 
the air, it shrinks considerably, and if in thick masses, numerous 
cracks are formed. The admixture of three or four volumes of sharp 
sand prevents this shrinkage by separating the lime paste into very 
thin layers, which fill the spaces between the grains of sand. The 
sand also gives the mortar a porous structure, which facilitates the 
penetration of the carbon dioxide during the hardening period. The 
interlacing crystals of calcium carbonate enclose the sand grains and 
join them together, thus increasing the hardness and strength of the 
mortar. This addition of sand also cheapens the mortar by increas- 
ing the mass obtained from a given amount of lime. " Fat lime " 
requires a much larger proportion, which is replaced in part by the 
impurities in " poor " lime. 

For a good mortar it is necessary that the lime be thoroughly 
slaked, in the proper quantity of water added all at once, or the 
product is apt to be granular and lumpy. The mass is covered with 


a layer of sand, or with boards or canvas, to retain the heat and 
moisture, and is not stirred w.hile slaking, but is allowed to swell and 
fall to powder without disturbance. Water is then added, and the 
paste allowed to stand for several days, or even weeks, well protected 
from the air, before being stirred up with more water for use in mortar. 

The first change noticeable in a mortar is the " set," which is a 
solidification of the mass, due to the loss of its water through evapo- 
ration or absorption by the bricks, etc. But it is not until the mass 
becomes nearly dry that the real hardening begins. This is very slow, 
since it progresses from without towards the interior of the mass ; 
and the surface layer of calcium carbonate first formed is but slowly 
penetrated by carbon dioxide from the air. The interior of thick 
walls will often show an alkaline reaction after the lapse of a century 
or two, but after twenty-five years the change is very slight under 
ordinary conditions. After several hundred years there appears to 
be a certain amount of combination between the silica of the sand 
and the calcium carbonate to form a hydrated silicate of calcium. 
This secondary reaction does not increase the hardness of the mor- 
tar. Hardening is a true chemical change, and should not be too 
rapid for the best results. In order to hasten the hardening of 
mortar and plastering in new houses, builders sometimes build coke 
or charcoal fires in open grates or baskets. But this is liable to 
cause uneven drying and excessive shrinkage, resulting in cracks or 
scaled places. In certain mortars hair or other fibrous material is 
added to increase the toughness, especially while wet. 

Since mortar does not harden until dry, it should never be used 
in damp places, such as foundations and cellars, nor in very thick 
walls. Sometimes it is mixed with some cement, increasing its 
strength and usefulness. When thoroughly hardened, good mortar 
is about as hard as limestone, and adheres firmly to the bricks or 
stones of the wall. 

Sand-lime bricks * made by mixing sand with about 8 per cent of 
slaked lime, moulding the mixture in a brick mould and subjecting 
the product to the action of steam at pressures of 130 to 150 Ibs. per 
square inch, for four or five hours, have come into general use for 
building purposes, especially for inside work. It is claimed that 
under the action of the steam at high pressure, the lime and sand com- 
bine to form calcium silicates, possibly similar to Portland cement. 
But this seems doubtful, as the temperature is not comparable with 

* Trans. Am. Ceramic Soc., 1903 ; 1911, 648. Jour. Soc. Chem. Ind., 1899, 48 ; 
1902, 1183; 1903, 421. 


that required for cement making. Possibly the bonding substance 
is a mixture of the three calcium silicates, but chiefly the meta-sili- 
cate (CaSiO 3 H 2 O). 


Cement consists of certain anhydrous double silicates of calcium 
and aluminum, which are capable of combining chemically with 
water, to form a hard mass. It differs from lime mortar in that it 
hardens while wet, does not require the presence of carbon dioxide 
for hardening, and is very insoluble in water. It is well adapted for 
use in moist places, or even under water, and since its hardening is 
simultaneous throughout the whole mass, and is quite rapid in most 
varieties, it finds extensive use in building operations. 

There are three general classes of cement : 

1. Those formed from certain volcanic tufas, or from artificial 
mixtures resembling these. Such cements generally need the addi- 
tion of lime before they display hydraulic properties, i.e. form in- 
soluble silicates when treated with water. In this group are the 
natural volcanic tufas, Pozzuolan, trass, and Santorin earth, together 
with blast furnace slags and certain coal ashes, which are occasion- 
ally used. 

2. Those which contain a large proportion of free lime, having 
been made by burning natural argillaceous limestones at a tempera- 
ture sufficiently high to drive off all the carbon dioxide, but not to 
fuse the product. These include " hydraulic limes " (p. 178) and 
Roman cements. 

3. Those prepared by burning an intimate mixture of clay or 
other alumino-siliceous material and powdered calcium carbonate, 
at a very high temperature, so that incipient fusion takes place in 
the mass. These constitute the Portland cements. 

Pozzuolanic cements are chiefly derived from volcanic tufas, 
found in Italy, near Naples (Pozzuoli), in the islands of the Grecian 
archipelago, and in Germany near Andernach on the Rhine. These 
tufas consisting of easily decomposable silicates have resulted from 
the action of volcanic fires, and need no further treatment than fine 
grinding and mixing with lime. Such cements are slow in hardening, 
but have considerable ultimate strength. Pozzuolan has been used 
since the time of the Romans, who were well acquainted with its prop- 

Blast furnace slag, high in alumina and silica, is now much used 
for Portland cement, limestone being the other ingredient. The 


melted slag is chilled and granulated by running into water and the 
sandy material is dried, ground fine, and mixed with the ground lime- 
stone. The dry mixture is then calcined to produce a clinker, which, 
after fine grinding, yields an excellent cement ; but it may be rather 
high (3 to 4 per cent) in calcium sulphide. 

Hydraulic limes have already been mentioned (p. 178). The free 
lime which they contain is sometimes slaked with just sufficient water 
to hydrate' the quicklime before the material is sold ; but not enough 
water should be added to set the cement. 

Roman cement is made by burning argillaceous limestone in kilns. 
It was first made in England by J. Parker, who patented a process 
for preparing it from the septaria nodules, consisting of clay and 
chalk found in the bed and along the banks of the Thames River. 
Later, the beds of clay limestones were used, but as there was much 
irregularity in the composition of these rocks, the product did not 
give satisfaction. But by careful selection of the material and 
proper mixing of different kinds of stone, the quality of cement pro- 
duced has been improved. These rocks are also found in France, 
Holland, and Germany, and in the United States. There are sev- 
eral deposits that are very pure and vary but little in the different 
parts of the bed. Nearly all these rocks contain a large percentage 
of magnesia, but this does not appear to injure the cement made 
from them. Roman cement was first made in this country in New 
York state, from a rock found on the banks of Rondout Creek and 
near the Hudson River, and is called " Rosendale," from the chief 
town in the district ; it still constitutes a large part of the natural 
cement made in this country. Another important region is on the 
Ohio River, near Louisville, the cements made there being known by 
the latter name. Pennsylvania, Illinois, Wisconsin, and Colorado 
also supply natural cement. 

The rock is broken into lumps about the size of a goose egg, in 
order to secure evenness in burning. The burning is done in con- 
tinuous kilns, as a rule, and the temperature must be carefully regu- 
lated, high enough to drive out the carbon dioxide, but not to fuse 
the rock. Then the rock is carefully ground and sifted. The finer 
the grinding, the better the product. In order to secure supposed 
uniformity in the product, it is often customary to mix rock from 
several beds, in the same kiln, but this is of doubtful benefit. 

The color of Roman cement varies greatly, from pale yellow to red- 
brown, and is due chiefly to the amount of iron and manganese oxides 
present. But there should not be great variations in the color of the 



products made from the same rock, as this indicates inequality in 

Roman cement is generally quick-setting, and hence is preferred 
by many engineers for work under water. It weighs from 50 to 
56 pounds per cubic foot. Its strength is inferior to Portland cement. 

The following analyses * are of typical cement rock : 







CaCO 3 
MgCO 3 . . 


1 11.38 
| 1.20 


| 1.12 

\ 1.07 
/ 2.46 

\ 3.03 
J 1.41 



SiO 2 


AL>o 3 

Na^O .... 

K 2 O 

H 2 O . . . 

Undetermined . . . 

Analyses of natural cements : 










A1 2 O 3 

1 16.70 

1 7K1 

\ 7.43 

Fe 2 O 3 


| 9.92 

> 7.51 







K 2 O 

\ * 

Na 2 O . . . 


| 0.80 

| 1.50 

CO 2 




CaSO 4 



H 2 O . 




Undetermined .< . . 


Portland cement is entirely an artificial product, but represents 
the most important branch of the cement industry. The first patent 
was taken out in England, in 1824, but the process extended in a 
few years to France and Germany. In the United States the manu- 
facture of this cement was begun in 1878, at Coplay, Penn., and 
the industry has become enormous. 

The materials used are calcium carbonate and clay rich in silica. 
Limestone and shale, or marl and clay, are used in this country and 

* W. A. Smith, Mineral Industry, Vol. I, 1892. 



Europe, and chalk and clay mud from the estuaries of the Thames 
and Medway rivers are preferred in England. But in any case the 
proportion of calcium carbonate to aluminum silicate must be con- 
trolled between tolerably narrow limits. 

The average composition of raw materials is shown in the follow- 
ing table : 





SiO 2 . 
AUO 3 . . 





Fe 2 O 3 



1 L5 


CaCO 3 





MgCO 3 





Alkalies, moisture, etc. 



A most thorough mixing of the ingredients is very essential. 
This is done in two ways : the " dry process " is used when the 
materials are hard (limestone and shale) ; and the " wet process " 
for soft materials containing much water (marl, chalk, and clay). 

The dry process is simple and cheap, but requires excessively fine 
grinding for uniformity of the product. The proper proportions 
(by weight) of shale and limestone are crushed together, and often 
lightly calcined to drive off moisture, and then ground in the Griffin 
mill or ball-mill so that 80 per cent passes through a 200-mesh sieve. 
This powder is usually fed directly to the rotary kiln, or may be 
pressed into bricks and burned in a shaft kiln or ring furnace. 

The wet process is carried out in various ways, according to the eco- 
nomic conditions prevailing in each locality. Usually the clay and 
chalk or marl are ground in edge-runners with heavy rolls, and water 
is added till enough is present (40 to 50 per cent) to make a slime or 
" slurry " which will flow or can be pumped. Sometimes a wash-mill, 
consisting of flat stones sliding over a smooth bed of stones or iron, 
propelled by a rotary shaft with arms, is used for this preliminary 
mixing and grinding ; or a disintegrator mill may be employed. 

The wet slurry is then pumped or run to buhrstones or tube-mills (p. 
188) and given a very thorough grinding. Often an intermediate mix- 
ing is given in tanks having rotary arms, or compressed air agitators. 

The next treatment of the slurry varies in different mills ; it may be 
settled or filtered and the excess water run away ; the mud may then 
be pressed in a brick machine, or dried on floors heated by waste heat 
from the kilns. The bricks or lumps are then sent to the kilns. In 



this country the slurry is generally run directly into the rotary kilns, 
the water evaporating in the upper third of the kiln, and the dried 
mass forming little balls or gravel-like lumps, which are thoroughly 
burned while passing through the rest of the kiln. 

Various kinds of kilns are used for Portland cement burning, the old 
periodic dome-shaped or shaft kilns being still in use in some places, 
although costly as to fuel and output. Continuous shaft kilns are 
more common, especially where labor is cheap and fuel high ; in 
America they have been displaced by the rotary kilns, which have 
much greater capacity. In shaft kilns, owing to the high temperature, 
the charge tends to stick to the furnace walls, unless careful attention 
is given during the burning. With rotary kilns, the amount of pow- 
dered coal consumed varies from 25 to 40 per cent of the weight of 
the cement produced. 

The Dietzsch two-storied kiln (Etageofen) (Fig. 76) is much used 
abroad for Portland cement. The charge, 
introduced at (A), fills the vertical shaft 
or pre-heating chamber, and absorbs the 
heat of the escaping combustion gases 
as it descends into the combusion cham- 
ber (B). The fuel, either coke or coal, 
is introduced through (D), while the ma- 
terial from (B) is raked down into the 
chamber (C), mixing with the fuel, and 
attaining there the highest temperature 
of the calcination. The burned material, 
mixed with the fuel ash, is withdrawn at 
(E). Air enters at (E), and passing 
through the hot clinker, arrives in (C) at 
a very high temperature, where it sup- 
ports the combustion of the fuel. The 
hot gases passing off through (B) and (A) 
serve to heat the charge before it arrives 

in (C). These kilns now are often worked with forced draught, and 
produce about 7 tons of clinker per ton of coal. 

Hoffmann's ring furnace (Fig. 77) is also much used in cement 
burning. This consists of an elliptical gallery built around a central 
chimney (A). The gallery is divided into 15 or 20 compartments 
(B, B) each having a door.(C) opening outside, a flue (D) leading to 
the chimney (A), and a wide opening (D) into the next compartment. 
Each flue has a damper, by which connection with the chimney (A) 

FIG. 76. 



may be opened or closed. The openings (E) between the compart- 
ments may be closed with a sheet-iron or heavy paper diaphragm, as 
will be explained below. If the door of the compartment on one side 
of the diaphragm be opened, and the damper of the flue (D) leading 
from the compartment on the other side of the diaphragm is also 
opened, while all the other doors and flues are closed, the draught of 
the chimney (A) will cause air to enter the open door and pass around 
the entire gallery, through each compartment in succession, and 
finally out through the open flue (D) to the chimney. In the roof 
over the gallery are charging holes (G), several being in each compart- 
ment, through which fuel is introduced. The furnace is run as 
follows : Assume that there are 14 compartments, as shown. Twelve 
compartments contain cement bricks, and their doors and chimney 

flues are closed. Sup- 
pose that No. 1 is 
being emptied, while 
No. 14 is being filled. 
The paper diaphragm 
closes the opening be- 
tween No. 13 and No. 
14, and the flue (D) of 
No. 13 is open to the 

chimney. Compartment No. 7 is at the height of combustion, while 
Nos. 6, 5, 4, 3, 2 contain bricks which have been burned. In Nos. 8, 9, 
10, 1 1, 12 are bricks to be burned. Cold air is drawn in through the open 
door of No. 1, and passing in order through Nos. 2, 3, 4, 5, 6 becomes 
heated by contact with the hot bricks in these compartments until, 
after passing through No. 6, which is still red hot, it arrives in No. 
7 at a very high temperature. In No. 7 the fuel is burning at a white 
heat, and the hot gases pass on through Nos. 8, 9, 10, 11, 12, 13, 
from which they escape to the chimney. By this passage of the hot 
gases through the compartments, the unburned bricks are heated, 
those in No. 8 being nearly red hot ; but as no fuel has been intro- 
duced into these chambers, combustion and white heat are confined 
to No. 7. When No. 14 is filled with green bricks, its doors are 
closed, and also the chimney damper of No. 13, while that of No. 14 
is opened, and the diaphragm transferred to the opening between No. 
14 and No. 1. Fuel is now introduced into No. 8, which becomes the 
combustion chamber, and the door of No. 2 is opened. The burned 
bricks in No. 2, having been cooled by the passage of cold air, are 
taken out, while No. 1 is being refilled. Thus the cycle of operations 


goes on, each compartment in turn being charged with fuel and made 
the combustion chamber. The temperature in that compartment 
which has just been filled is only high enough to dry the green bricks. 

This furnace is very economical of fuel, one ton of soft coal burn- 
ing 6j tons of clinker, but it requires much labor. The bricks must 
be accurately piled in order that open channels may be left beneath 
the charging holes for the fuel, which is thus made to burn in a column 
extending from the floor to the top of the furnace. Usually one com- 
partment is emptied each day, and the fire is moved forward. 

Revolving furnaces (Fig. 126, p. 598) are largely used for cement 
burning, and have greatly advanced the Portland cement industry in 
America. These kilns incline about 5 to 8, are 60 to 220 feet long, 
6 to 8 feet in diameter, and are lined with fire-brick. The fuel is 
usually powdered coal, blown in at the lower end of the furnace by 
an air blast ; in rare cases oil or natural gas may be used. The mix- 
ture of powdered materials (or slurry, in the wet mixing process) 
enters at the upper end, and is thoroughly calcined during the two 
or three hours' passage through the furnace and sintered by the 
heat into little balls or gravel-like lumps. The hot clinker is dis- 
charged into an elevator which carries it to the iron coolers, in which 
air conies in contact with the mass. A little water is sometimes 
sprinkled into the elevator buckets to " cure " the clinker and insure 
rapid cooling, which assists materially in the grinding. 

The rotary kiln is cheap in its labor account but even the large 
ones are wasteful of fuel ; probably the lowest consumption is about 
75 Ibs. of coal per barrel of clinker ; the Dietzsch and Hoffmann kilns 
use about one-third this amount of fuel, but require much higher ex- 
pense for labor, which probably explains the more general use of 
these types in Europe, where fuel is costly and labor relatively cheap. 
Much depends on the calcination temperature, which should reach 
1400 to 1600 C. (incipient fusion), and there is considerable shrink- 
age ; well-burned clinker is semi- vitrified, hard, and of greenish-black 

The clinker is then ground in various types of mills (see below), 
to such fineness that not over 7 or 8 per cent is retained on a sieve 
having 100 meshes to the linear inch. Only the very fine dust is 
considered of value in cement. 

The ball-mill (Fig. 78) consists of a cast-iron drum (D), contain- 
ing numerous chilled-iron or steel balls (B) of different sizes. The 
clinker is powdered by the rubbing and pounding of the balls as the 
drum rotates. The dust passes through the perforated plates (P) 



and falls on the fine sieves (S). The coarse particles retained by the 
screens return to the interior of the drum through the openings 
(C, C) for further grinding. 

FIG. /d. 

The tube-mill (Fig. 79) is a horizontal iron tube about 20 feet 
long by 5 feet in diameter, rotated some 25 times a minute by the 
gears (G). The tube is half full of smooth quartz pebbles, about the 
size of goose eggs ; these are retained in the tube by a screen (S) at 
the outlet end, through which the ground cement passes and is dis- 
charged through (T). The pebbles are slowly ground up with the 
cement and a few new ones are added with the clinker at regular 
intervals. The speed of rotation and the rate of feed of material 
through the trunnion by the screw (P) determine the fineness of the 

j ^ 

FIG. 79. 

product. These mills require considerable power and have rather 
small capacity, but any degree of fineness can be obtained, and the 
repairs are very moderate. They work equally well on dry or wet 
materials and are often used to mix and grind slurry. 

The Hardinge conical mill (Fig. 80) is a modification of the tube- 
mill, having somewhat greater efficiency. 

The Griffin mill is a steel roll, weighing about 400 pounds, revolv- 



FIG. 80. 

ing on a vertical shaft with a gyratory motion, and pressing by cen- 
trifugal force against a steel ring. It has great capacity, and will 
grind so that about 90 per cent of the product 
passes a 100-mesh sieve ; but the repair account 
is rather large. 

The Fuller-Lehigh mill consists of a hori- 
zontal ring with a ground inner surface, against 
which several heavy steel balls are made to 
revolve at a speed of about 160 r. p. m. The grinding effect is due to 
the centrifugal force developed by the whirling balls. 

The Kent mill (Fig. 81) consists of three rolls rotating at high speed, 
within a movable circular ring or die, 
against which the crushing takes place. 

The constitution of Portland cement 
has been much studied, and various views 
are held as to the proper proportions of 
the ingredients. Le Chatelier states * 
that the ratio of the equivalents of lime 
and magnesia to the silica and alumina 
should not exceed a maximum equal to 
three, while that of the total silica, minus 
the combined iron and alumina, should not be less than a minimum 
equal to three ; thus : 

FIG. 81. 

CaO + MgO ^ CaO + MgO 


Si0 2 + A1 2 3 

Si0 2 - (A1 2 3 + Fe 2 3 ) = 


According to Michaelis, the ratio of lime to the acid constituents 
should fall between 


SiO 2 A1 2 O 3 

> 1.8 and 


SiO 2 A1 2 O 3 Fe 2 O 3 

< 2.2. 

Newberryf holds that the proportion of lime, by weight, should 
be to the silica and alumina as shown thus : 

Lime = (SiO 2 X 2.8) + (A1 2 O 3 X 1.1). 

The composition of good commercial cements, however, shows 
some variation, and no definite formula can be assigned to them. 
The following are typical : 

* Trans. Am. Institute Mining Engineers, 22 (1893), 15. 
t J. Soc. Chem. Ind., 1897, 887. 



SiO 2 21.05* 22.80* 19.78* 21.50f 22.04f 21.25f 

A1 2 3 

8.95 6.49 8.21 6.60 6.45 6.16 

Fe 2 O 3 4.40 4.31 

CaO 61.30 61.10 

MgO 1.37 .47 

SO 3 1.28 1.39 

Alkalies 68 .30 

H,O, CO 2 70 2.40 









3.41 3.85 

60.92 62.69 

3.53 3.00 

2.25 1.50 



99.73 99.26 99.68 95.38 98.60 98.95 

The cause of hardening of cement has been explained in various 
ways. Le Chatelier J holds that during the burning a tricalcium sili- 
cate (Ca 3 SiOs) is formed by reaction between the clay and the lime, 
and at the same time, some calcium aluminate and ferrite are formed, 
besides mono- and di-calcium silicates. By the action of water on 
the tricalcium silicate, hydrated monocalcium silicate and calcium 
hydroxide are formed : 

1) 2 Ca 3 SiO 5 + 9 H 2 O = (CaSiO 3 ) 2 5 H 2 O + 4 Ca(OH) 2 . 

Then the calcium hydroxide, water, and calcium aluminate may react 
to form hydrated basic calcium aluminate : 

2) Ca 3 Al 2 O 6 + Ca(OH) 2 + 11 H 2 O = Ca 4 Al 2 O 7 12 H 2 O. 

The formation of the hydrated basic aluminate [(CaO) 4 A1 2 O 3 12 H 2 O] 
is supposed to influence the setting of the cement, but the hardening 
is ascribed to the first reaction. Richardson takes the view that 
Portland cement clinker is largely composed of alit, a solid solution 
of tricalcic silicate in tricalcic aluminate; and that the setting is 
due to the decomposition of alit, with formation of crystals of cal- 
cium hydroxide. Hydration of the silicates and alumina tes is not 
thought to add to the strength of the cement after setting, but the 
crystallization of the calcium hydroxide binds the mass together. 

Portland cement is usually slower in setting than Roman, but 
when the hardening has begun, it progresses more rapidly with the 
former. There is very little increase of hardness after six months. 
Portland cement is more durable than Roman under most conditions, 
and is generally stronger. It forms a denser and heavier powder of 
a greenish gray color, but when hardened has a drab shade resembling 
the color of the stone quarried at Portland, England, and used much 
for building in that country ; hence the name. Variations in color 

* English. f American. 

J Annales des Mines, 1887, 388. J. Soc. Chem. Ind., 1888, 567, 847. 
dustrie Zeitung, 16 (1892), 1032. Chemiker Zeitung, 1892, Ref. 342. 
Proc. Assoc. Port. Cement Mfgr., 1905, June 15. 



of the same brand of cement may show changes in quality ; if under- 
burned, it is generally yellowish. The weight per cubic foot varies 
from about 70 to 90 pounds ; the finer the grinding, the less the weight. 
But as a rule heavy cements are preferred by builders, as they are sup- 
posed to be more thoroughly burned ; they are, however, slow in setting. 

The testing of cement is generally the work of the engineer. Chemical 
analysis alone is of small use in determining its properties, and physical 
tests are usually more satisfactory. Committees from the American 
Society for Testing Materials,* and from the American Society of Civil 
Engineers,! and other engineering associations have adopted " Standard 
Specifications " and " Methods of Testing." The tests recommended, 
are for : 

(a) Specific Gravity. (6) Fineness. (c) Time of Setting. 
(d) Tensile Strength. (e) Soundness. 

The specific gravity of the cement dried at 100 C. shall not be less 
than 3.10. The determination is made in Le Chatelier's apparatus, 
using naphtha of 62 Be., or kerosene. 

In testing for fineness, not more than 8 per cent by weight may 
remain on the No. 100, nor more than 25 per cent on the No. 200 sieve. 
Use circular sieves, 20 centimeters in diameter, with woven cloth of 
brass wire 0.0045 inch and 0.0024 inch in diameter respectively, for 
the sieves. Fifty or 100 grams cement, dried at 100 C., are to be used 
for the test. 

Time of Setting. The cement shall develop the initial set in not 
less than 30 minutes, and should set hard in not less than one hour nor 
more than 10 hours. The test is to be made with the Vicat needle, 
which is a movable vertical rod, with a plate on the upper end and a 
short cylinder, 1 millimeter in diameter, at the bottom, the whole sup- 
ported in a frame. The rod, plate, and foot-piece weigh 300 grams. 
A paste of cement and water is put into a frame under the needle point, 
and the depth to which the needle sinks in the soft mass is noted. The 
set has commenced when the needle ceases to penetrate to within 5 
millimeters of the glass plate on which the paste rests, and is terminated 
when the needle no longer enters the mass. 

Time of setting determinations are not exact, and vary with the 
quantity of water used in the mortar, the temperature of the water 
and of the air, and the amount of working the mortar may have re- 
ceived during the moulding for the tests. If the set begins in less than 
half an hour, the cement is called " quick-setting," and is desirable for 
work under water. Slow-setting cement requires more than half an 
hour for the set to begin; this is better for most purposes and may be 
mixed in larger quantities. High temperatures hasten the set of the 
cement, while a larger proportion of water induces slower setting. The 

* Proc. Am. Soc. Testing Materials, 1904. 

t Trans. Am. Soc. Civil Engineers, 1903 ; amended 1904. 



addition of not more than 2 per cent of calcium sulphate, as gypsum or 
plaster of Paris, retards the set materially. Larger quantities may hasten 
the set. 

For ordinary control work, the time of setting may be determined 
sufficiently closely by the " normal needle," devised by Gilmore. Two 
of these are used : one is a wire one-twelfth of an inch in diameter and 
loaded with a weight of one-quarter of a pound ; the other wire has a 
diameter of one twenty-fourth of an inch and carries a weight of one 
pound. The cement, mixed to a stiff paste with water, is formed into a 
pat, one-half an inch thick, and the time noted until no impression is 
made upon it by the point of the first wire. This is the beginning of the 
'* set." When the second wire will not penetrate, the set is ended. 

The tensile strength is determined on a briquette, shaped like an 
hour-glass, and having at the narrow portion a section exactly one inch 
square. The minimum requirements shall be within the following 
limits : 



For "neat " cement 
(i.e. without sand) 

Sand briquettes (1 part 
cement, 3 parts sand) 

24 hours in moist air 
7 da. ( 1 da. in moist air ; 6 da. in 
28 da. (1 da. in moist air ; 27 da. in 
7 da. (1 da. in moist air ; 6 da. in 
28 da. (1 da. in moist air ; 27 da. in 

150-200 Ib. 
450-550 Ib. 
550-650 Ib. 
150-200 Ib. 
200-300 Ib. 

For the sand briquette, a natural sand, obtainable at Ottawa, Illinois, 
is recommended, but many engineers prefer a standard sand made by 
pulverizing pure quartz. The sand is sifted and that portion used which 
passes a No. 20 sieve and is retained by a No. 30. The briquettes must 
be carefully made to secure uniform results. The cement is mixed with 
water at about 70 F., filled into bronze moulds, pressed down well, and 
smoothed off evenly. This is done on a slate or glass plate to prevent 
absorption of moisture. When set, the briquette is removed and placed 
in a moist-air closet for 24 hours. It is then kept in water until the test 
is made, when it is placed in the jaws of a machine, which applies a gradu- 
ally increasing tension at the rate of 400 pounds per minute. The num- 
ber of pounds necessary to fracture the briquette is read on a graduated 
scale beam. The average of three tests (of each neat and sand-mixture) 
is usually taken as the tensile strength. 

The less water used in the cement mortar, the higher the strength, 
as a rule, especially in the short time tests. For neat cement, the water 
may vary from 14 to 24 per cent ; for sand briquettes, about 12 per cent 
of the total weight of the sand and cement. The cement and sand should 
be well mixed, dry; then wet out, and mixed with water in about 2^ 
minutes, and filled into the moulds at once. 


Compression tests are made with small cubes of the cement. This 
test should show at least ten times the tension resistance. This being 
difficult to manage, the tensile test is employed instead usually. 

Soundness tests are made upon pats of neat cement, 3 inches in 
diameter, one-half inch thick at the centre and tapering to a thin edge. 
These are kept in moist air for 24 hours ; then one is exposed to air at 
ordinary temperature for 28 days ; another is kept in water at 70 F. for 
28 days; a third is exposed to steam in a loosely covered vessel for 5 
hours. All of the pats must show no signs of checking, cracking, dis- 
integrating, nor distortion. Faija's test is often used. This consists 
in placing the test piece in a moist atmosphere at 100 to 105 F., for 6 
hours or more, till well set ; then it is immersed in water at 115 to 120 F. 
for the remainder of 24 hours. 

Expansion or " blowing " is shown by swelling, cracking, or disin- 
tegration of the cement after setting. This is generally supposed to 
be caused by excess of free lime, or by poor burning, the heat not having 
been enough to combine the lime with the silica and alumina. The free 
lime slakes after the cement has set, and the expansion causes disin- 
tegration. Magnesia in Portland cement has been thought to cause 
unsoundness, but up to 4 per cent (as MgO) appears to be harmless, and 
is allowed by the Standard specifications. . 


Plaster of Paris is made by heating gypsum (CaSO4 2 H 2 O) until 
about three-fourths of its water of crystallization is driven off. The 
process, called burning, is carried on in muffle furnaces, kettles, or 
rotary cylinders. Direct contact with the fuel is avoided, lest the 
carbonaceous matter may reduce some of the calcium sulphate to 
sulphide ; nor should the flame come in contact with the gypsum, 
but only the hot gases. Boiler-iron kettles, having a stirring device, 
are much used, the charge being 5 to 7 tons of finely ground gypsum, 
which requires about three hours to calcine. Rotary calciners are 
being adopted because of their economy ; these are iron cylinders set 
at a slight incline, and heated externally by the flame from a furnace. 
Gypsum, crushed to small size, is fed continuously into the cylinder ; 
the calcined material is friable and easily ground fine. 

Gypsum contains about 21 per cent of water of crystallization, 
while plaster retains 4 to 7 per cent. Moisture begins to escape at 
about 100 C., but the most favorable temperature for calcining is 
around 145 C. ; if heated to 200 C., all the crystal water is expelled, 
and the product will combine with water very slowly, the property 
of rapid setting being lost. Thus the limits of heating are very 
narrow and much care is needed in the process, 


When mixed with water, plaster of Paris forms a paste which 
soon hardens or " sets," owing to a recombination of water with the 
burned plaster, to form hydrated calcium sulphate. The theory of 
this setting has been explained by Le Chatelier.* The composition 
of the plaster is essentially (CaSO^ H2O, a salt which is soluble, 
and part of which dissolves in the water used in mixing. But as 
soon as it dissolves, a combination between it and some of the water 
takes place, forming CaSO4 2 I^O ; this, being much less soluble 
than the monohydrated salt, at once begins to crystallize from the 
solution, forming a network of crystals. Then more of the plaster 
dissolves, becomes fully hydrated, and crystallizes out, increasing 
the solidity of the " set " by the interlacing of new crystals with 
those already formed. Thus the cycle of reactions goes on until the 
plaster is fully hydrated. 

The theoretical quantity of water necessary to set plaster is about 
18 per cent of its weight ; but in fact, from 30 to 35 per cent is generally 
used. Excess of water renders the mass more plastic and retards the 
setting. Large excess causes disintegration of the plaster, if left in 
contact with it for some time after setting, owing to the solution of 
some of the crystallized calcium sulphate. Since plaster " sets " 
very rapidly with water, a " retarder," such as glue, dried blood, or 
vegetable gums, marshmallow root, or fine sawdust, is often added. 
These colloidal bodies probably decrease the solubility of the calcined 
plaster and retard the hardening for half an hour or more. 

Plaster expands slightly while setting, and hence is valuable for 
making casts and reproductions. It is used for interior decorative 
work and also as a cement for joining glass and metal ware. The 
surface of plaster after setting is rather soft, and if it is desired to 
increase the hardness, this may be done by mixing alum, borax, or 
tartaric acid with it, or by adding some alcohol to the water with 
which the plaster is mixed. However, these substances retard the 
setting. By painting or dipping plaster casts in melted wax, paraffine 
or stearin, or in solutions of these in petroleum ether, the pores are 
filled and the surface is made smooth, so that dirt will not adhere 
and the articles may be washed. When treated with a solution of 
barium hydroxide, the surface of the plaster is coated with barium 
sulphate and rendered insoluble. If plaster is mixed with a solution 
of glue or size, the material called " stucco " is obtained. 

* Comptea Bendus, Vol. 96, 717, 1668. 



Die hydraulische Morter. Michaelis, 1869. 

A Practical Treatise on Limes, Hydraulic Cement, and Mortars. Q. A. 

Gilmore, 1874. 

Chemische Technologic der Mortelmaterialien. G. Feichtinger, 1885. 
Recherches experimentales sur la Constitution des Mortiers hydrauliques. 

Le Chatelier, Paris, 1887. 
Fabrication et Controlle des Chaux hydrauliques et des Ciments. H. 

Bonnami, Paris, 1888. (Gauthier-Villars et Fils.) 
Zement und Kalk. Rudolf Tormin, Weimar, 1892. (B. F. Voigt.) 
A Manual of Lime and Cement. A. H. Heath, London, 1893. (Spon.) 
Annales des Mines: XI (1887), 388-465. H. Le Chatelier. 
Cements, Limes and Plasters. Edwin C. Eckel, New York, 1907. 
The Modern Manufacture of Portland Cement. Percy C. H. West, 1910. 
Portland Cement. Richard K. Meade, 2d ed., Easton, Pa., 1911. 
Journal of American Chemical Society : 1894, 161. T. B. Stillman. 
Journal of the Society of Chemical Industry : 

1886, 188, 199. 1891, 927. 1897, 887. 1910, 1107. 
Trans. Am. Inst. Min. Eng., 27, 508. P. Wilkinson. 
Transactions of the American Society of Civil Engineers : 

1877. W. F. Maclay. 1885. E. C. Clarke. 

1885, 1903 ; 1904. Report of Committee on Cement Tests. 

1893. Max Gary. 


Glass is an amorphous, transparent, or translucent mixture of 
silicates, one of which is always that of an alkali. The usual silicates 
employed are those of potassium, sodium, calcium, and lead ; the sili- 
cates of heavy metals occur in the colored glasses. Glass is not 
readily decomposed by water or acids (excepting HF). Its behavior 
towards solvents generally tends to show that it is a mixture of 
silicates, rather than a definite compound. 

Most simple silicates and mixtures of them are difficult to fuse, 
and when cooled after fusion, have a crystalline structure ; but the 
alkali-lime and alkali-lead silicates fuse easily, and are generally 
amorphous after fusion. Silicic acid is capable of forming a number 
of salts of varying acid content and of approximately equal stability. 
Hence from a fusion of the silicates of two or more metals, the tendency 
of any particular compound to crystallize is small, and even if a 
crystal centre is formed, growth is very slow, because of the viscosity 
of the medium and the major part of the liquid consisting of dissimilar 
molecular structures. Hence such a mass can be supercooled, the 
viscosity progressively increasing, and the melt becoming first plastic 
and finally rigid. Such a supercooled liquid is called a glass ; it has 
no melting point, but softens progressively on heating, and if kept 
thus for some time crystallization, technically called " devitrification," 
will begin, causing a white or porcelain-like appearance. The alkali- 
lime and alkali-lead silicates are rigid at ordinary temperatures, in- 
soluble in most liquids, decrease in viscosity slowly when heated, so 
that they are plastic and workable over a large temperature range, 
and devitrify very slowly. These properties give them their technical 

Since the glasses are complex, homogeneous mixtures, all of their 
properties can be progressively changed by a progressive modification 
of composition, and use of this is made in the arts. 

Soda-lime glass approaches Na2O, CaO, 6 SiO 2 , and lead glass, 
IQO, PbO, 6 SiO 2 ; but it may vary so much that the formula becomes 
5 K 2 O, 7 PbO, 36 SiO 2 . Of course potash may be substituted for soda, 
or vice versa, in either kind, while the relative proportion of the several 
ingredients may vary between quite wide limits. But as a rule, the 
higher the percentage of silica, the harder, more difficultly fusible, 
and more brittle the glass. Increase of alkali makes it softer, more 


GLASS 197 

fusible, and less capable of resisting atmospheric changes and chemical 
reagents. Increasing the percentage of lime decreases the fusibility 
and renders it harder, but not so brittle as in the case of high-silica 
content. If the alkali used be mixed soda and potash, a more fusible 
glass is obtained than from either alone. Part of the lime or lead may 
be replaced by oxides of other metals, e.g. of iron, manganese, cobalt, 
copper, barium, zinc, tin, arsenic, etc., and this is generally the case, 
to some extent, in common glass, and to a greater degree in colored 
glass. Aluminium oxide may replace some of the silica ; the former 
is often present in considerable amounts, and renders the product 
tough. Certain fluorides, e.g. calcium fluoride, also enter into the 
composition of some varieties. Besides the above-named oxides, cer- 
tain borates and phosphates are occasionally used, to replace a part 
of the silica in glass manufactured for various optical and chemical 
purposes ; these usually contain zinc or barium also. The well-known 
" optical glass," made in Germany, contains both zinc and boron. 

Technically, two kinds of glass are recognized : lime glass and 
lead glass. The alkali used may be soda, or potash, or both. Lime 
glass is most common and generally useful. It is cheaper, harder, 
more resistive, and less fusible than lead glass ; the latter has greater 
lustre and brilliancy, is heavy and expensive and is used chiefly for 
cut ware and for optical purposes. 

The essential materials for glass making are silica, an alkali, and 
lime or lead. Silica was formerly derived from quartz or flint ; but 
this is now only used for a particularly fine quality. It is heated to 
a red heat, and dropped into water, and the friable mass so formed 
is powdered in a mill. Quartz sand and soft quartzites are the usual 
sources of silica, and numerous deposits are worked in different coun- 
tries. Sand of great purity is found in Germany, near Aix-la-Chapelle, 
and at Nivelstein ; in France, at Fontainebleau ; in Belgium ; in Eng- 
land ; and in Australia. In the United States, extensive beds are 
worked in Berkshire Co., Mass., and in Pennsylvania, along the Juniata 
River. The Berkshire deposit is a soft white sandstone, which, when 
crushed, yields sand which is from 99.6 to 99.8 pure SiO 2 . The Juniata 
stone is slightly yellow in color, and the sand is from 98.8 to 99.7 pure 
SiO 2 . The most troublesome impurity in sand is iron ; for white glass, 
there should never be more than 0.5 per cent Fe 2 O 3 . 

Alkali is derived from the carbonate or sulphate of soda or potash, 
and these also must be free from iron. Carbonate fuses more readily 
with the sand than does sulphate, but since the latter is cheaper, it 
is much used. It is essential to mix carbon in some form with the 


sulphate, to assist in reduction. For better grades of glass, charcoal 
dust is used, but for common glass, powdered coal is the reducing 
agent. The exact nature of the reaction with sulphate appears some- 
what uncertain : 

Na 2 SO 4 + SiO 2 + C = Na 2 SiO 3 + SO 2 + CO.* 

For lead glass, sulphates are not generally used, since some sodium 
sulphide is formed by the reduction of the sulphate, and this reacts 
with the lead, forming lead sulphide, which darkens the glass. 

Attempts to use salt directly in the glass furnace, as a source of 
alkali, have not proved satisfactory. It is quite volatile at the tem- 
perature of the furnace, and the presence of air or steam is necessary 
for its decomposition by the silica. 

For potash, crude pearlash may be used ; but in the better grades 
of glass the refined pearlash is employed. Sulphate of potassium is 
difficult to reduce, and is not much used. 

Lime is derived from chalk or limestone. For very fine glass, pure 
marble dust, as free as possible from iron, is employed. For common 
grades, less pure limestone is used. It may contain a high percentage 
of silica and considerable alumina, but magnesia or iron in large 
amounts is objectionable. Magnesia makes the glass hard and infu- 
sible. In cheap glass, limestone is sometimes replaced in part by fel- 
spar, porphyry, or granite. Carbonates of both alkali and lime are 
advantageous in the glass mixture, since, as the mass fuses, the escap- 
ing bubbles of carbon dioxide serve to stir up and mix the ingredients 
more thoroughly. 

Lead is added as litharge (PbO), or red lead (Pb 3 O 4 ). The latter 
is preferred, since the oxygen liberated from it is thought to assist in 
decolorizing the glass by oxidizing the iron ; it also prevents reduction 
of metallic lead. It is essential that the litharge and red lead be free 
from copper and silver. 

Besides the above requisites, it is customary to employ other in- 
gredients in every glass mixture, to assist in the decolorization or 
fusion. The commonest decolorizing material added is pyrolusite 
(binoxide of manganese, MnO 2 ). Iron, when in the ferrous condition, 
imparts a green color to glass ; but when in the ferric state, it is much 
less troublesome, since it only gives a pale yellow color. By the 
oxidizing action of the pyrolusite, ferrous iron is converted to the 
ferric condition ; moreover, the silicate of manganese has a violet or 
pink color, and so helps to neutralize the green. Only a very small 

* Lehrbuch der technischen Chemie, H. Ost, 8 te Auf. 257. 

GLASS 199 

percentage of pyrolusite should be thus used. The remedy is not a 
permanent one, and if the glass is exposed to the sunlight for a long 
time, it develops a violet shade, as may often be observed in the 
window panes of old houses. 

Arsenious acid (As2Oa), or nitre (NaNOa), is often added to the 
materials for colorless glass. The former is reduced to metallic 
arsenic, which volatilizes. It affords a very clear and lustrous glass. 
Zinc oxide is often used to decompose any sodium sulphide, which 
would give a yellow tinge to the product. 

In common bottle glass and other cheap grades, where color is 
no objection, blast furnace slag is often used. This generally needs 
the addition of soda, to render it more fusible and plastic. 

The formulae for glass mixtures vary much in the different fac- 
tories, not only because of variations in the composition of the glass 
produced, but also because the materials are of different degrees of 
purity. These are often empirical recipes, not based on analysis of 
the raw materials. 

The fuel for glass making is an important item. A quick-burn- 
ing material, yielding a long flame, without smoke or soot, is desir- 
able. For fine grades, wood is still employed in some places, but 
good coal is now most common. In this country the discovery of 
natural gas had a great influence on the glass industry, and within a 
few years most of the larger plants were moved into the gas territory, 
Pittsburgh becoming the centre of the manufacture. With the decline 
of the natural gas supply, producer gas (p. 41) or oil has been substi- 
tuted as fuel. Gas is an ideal fuel for this purpose, since it is clean, 
easily managed, and gives a regular heat. It is generally employed 
in regenerative furnaces (Fig. 23, p. 44). Crude petroleum, or the 
residuum from kerosene distillation, is much used and is a good fuel. 
Whatever the mode of heating, only the flame and hot gases should 
come in contact with the pots or their contents. 

There are several forms of glass-furnaces. The common pot fur- 
nace has the pots placed in a circle around a central opening in its 
bed, through which the flame and hot gases come up 'from the grate, 
which is below the hearth. The furnace is roofed with a rather flat 
arch, which deflects the flame down upon and around the pots. When 
open pots are used, it is essential that no soot or smoke enter the fur- 
nace, and much care is necessary in firing. In some forms, the fuel is 
introduced by mechanical means from beneath the grate, so that the 
fire burns on top of the pile of coal. This prevents the entrance of 
cold air into the furnace, and also consumes all smoke. 



The Boetius furnace (Fig. 82) is used abroad, and is best adapted 
for closed pots. Coal or coke is charged at (A), and air enters through 
(B, B). The flame passes through (C) into the upper compartment (D), 

containing the pots. The prod- 
ucts of combustion escape 
through (E, E). 

Siemens gas furnace (Fig. 
23, p. 44) is much used because 
of its economy of fuel, both as 
a pot furnace and as a tank- 
furnace. The last named is 
more economical where a large 
quantity of one kind of glass 
is to be made. It replaces the 

FIG. 82. . ' 

expensive and tragile pots by a 

single large deep hearth or tank, at one end of which the raw materials 
are continually introduced, while the glass is withdrawn at the other. 
Figure 83 shows a plan and elevation of a tank-furnace, in which the 

FIG. 83. 

batch is introduced at (A). The gas-flame issues from (C, C) and 
plays over the surface of the charge. The batch (B) soon fuses and 
the liquid mass flows towards the opposite end of the tank. At (F) 
are elliptical " floaters " of fire-clay, one end of which rests in recesses 
in the wall, while the free ends meet in the middle of the furnace. 
The current of melted glass flowing towards (D) constantly presses 
these floaters together and prevents their separation. The liquid 
mass thus passes under the floaters and collects in the compartment 



(D), from which it is withdrawn through the openings (E, E). At 
(B) the temperature is very high, and as the glass flows slowly towards 
(F), the refining takes place. In (D) the temperature is lower and 
the glass has cooled sufficiently for working. The impurities, rising 
to the surface during the melting and refining, are retained by the 
floaters so that the glass in (D) has a clean surface and is free from 
bubbles. Small rings of fire-clay may be kept floating on the glass 
near the working doors (E, E) ; by dipping the glass from the centre 
of these rings, it is obtained free from any impurities which may be 
on the surface of the melt in (D). A typical furnace of this kind may 
be about 75 feet long by 16 feet wide and 5 feet deep, to the level of 
the doors (E, E). 

Glass-furnaces must be made from very refractory materials. 
The dome and arches are usually silica, or Dinas bricks, or ganister, 
but the bed is generally fire-clay, as this is less attacked by the con- 
tents of a pot when one breaks. The life of a furnace is very uncer- 
tain, but may be several years. If allowed to ^_ 
cool, it is generally necessary to reline it before CjSilMh 4B58 
starting again. 

Pots for glass making are very carefully con- 
structed, only the best material being used. 
The breaking of a pot in the furnace is a seri- 
ous matter, often resulting in the loss of the 
glass and there is more or less loss of time. 

Glass-pots are of two kinds, open and closed. 
Open pots (Fig. 84) are circular vessels, about as wide as they are 
deep, i.e. from 3 to 5 feet, and usually slightly broader at the top 

than on the bottom. 
They are preferred for 
a quick melt, and 
are generally used for 
glass which contains 
no lead. 

Closed pots (Fig. 
85) are usually longer 
in one direction, and 
are about 5 feet by 3j feet, by 4 feet high. The neck of the opening 
is built into the wall of the furnace in such a manner that neither 
flame nor fire gases can come into contact with and injure the glass, 
and consequently cheaper fuel may be used ; but these pots heat 
more slowly than do open ones. They are always used for lead glass. 

FIG. 84. 

FIG. 85. 


Clay rings are sometimes placed in the pots, so that the glass 
may be withdrawn without contamination from the floating impuri- 
ties. Sometimes a partition is constructed across 
the pot (Fig. 86), the raw materials being intro- 
duced and melted on one side, and the refined 
glass, free from impurities, having passed under 
the partition, is worked out on the other side. 

The material of the pots is fire-clay ; but the 
necessary degree of plasticity, with the required 
infusibility, are possessed by but few clays. To 
avoid excessive shrinkage when the new pot is 
heated, a large proportion of burned clay from old pots, entirely free 
from any adhering glass and ground to a coarse powder, is mixed 
with the new clay. The mass is then moistened and well kneaded 
by treading, and is then allowed to stand and " age " for a long 
time, to increase the plasticity. The pots are built up by hand, the 
bottom being formed first, and the sides constructed on it. The clay 
is laid on in small lumps, and each lump is carefully pressed into 
place by the workman before another* is added. From three to five 
inches is usually added to the height of the pot each day. When 
finished, it is allowed to stand in a room at constant temperature and 
protected from draughts of air, for several months, to dry thoroughly. 
In order to prevent too rapid drying, which might cause cracking, it 
is generally covered with canvas or paper for the first few weeks. 

Before placing it in the glass-furnace, a new pot is heated very 
slowly in a special furnace, until it is brought up to the temperature 
of the former, into which it is then transferred, while still hot, through 
an opening in the wall. The wall must be taken down, the broken 
pot removed, and the new one introduced, without allowing the fur- 
nace to cool ; hence the operation is difficult, and requires much skill 
on the part of the workmen. Once introduced, a pot is kept in con- 
stant use, and never allowed to cool ; for, if it should, it would crack 
when heated again. Its life is very uncertain, but a good one will 
sometimes last for months. The first charge in a new pot is broken 
glass (cullet), which forms a glaze over the surface and protects it 
from the solvent action of the melted raw materials. 

The general process of glass making is as follows : The finely 
ground raw materials are thoroughly mixed, sometimes by regrinding 
the mixture or " batch." The batch is shovelled into the pot, together 
with a certain amount of broken glass called " cullet " ; this melts at a 
comparatively low temperature, and thus assists in liquefying the rest 



of the charge. More of the batch is added, until the pot is filled to 
the desired height with the fused mass ; then volatile substances, such 
as arsenious acid, used in decolorizing the glass, are added. 

During the melting, much gas (CO 2 , SO 2 , and O) escapes, and the 
bubbles rise through the melt, stirring it and causing frothing. A 
considerable amount of the alkali and other constituents volatilize. 
The reactions involved are variously written by different authorities : 

(a) 1) Na 2 CO 3 + CaCO 3 + 2 SiO 2 = Na 2 Ca(SiO 3 ) 2 + 2 CO 2 .* 

2) 2 Na 2 SO 4 + 2 SiO 2 + C = 2 NasSiOg + CO 2 + 2 SO 2 .* 

3) Na 2 SiO 3 + CaCO 3 + SiO 2 = Na 2 Ca(SiO 3 ) 2 + CO 2 .* 

(b) 2 Na 2 SO 4 + 6 Si0 2 + C = 2(Na 2 O, 3 SiO 2 ) + 2 SO 2 + CO 2 .f 

When the melt has come to a state of quiet fusion, the tempera- 
ture is generally raised somewhat, and the liquid glass allowed to 
stand for a time. This is called " refining," and its object is to form 
a homogeneous mass, free from bubbles and bits of uncombined silica 
or other matter. The scum which collects is skimmed off; it is 
called " glass gall," and consists of undecomposed sulphates and chlo- 
rides of lime and alkali, alumina compounds from the pot, and various 
other impurities. If too little carbon is used in the batch, the melt is 
covered with a layer of fused sodium sulphate ; this is known to the 
workmen as " salt water." Samples of the glass are examined during 
the refining, and these determine the exact time of heating. After 
refining, the glass is too liquid to blow, or to work to advantage, and 
is cooled until it becomes pasty. 

The quantities of materials used in the batch, for some typical 
glasses, are shown in the following table : 



Na 2 SO4 



MnO 2 

Pb 3 04 

K 2 COa 



French Plate 



















Lead flint 





Bottle glass 






(Green glass) 

Glass is known under various names in commerce, according to 
the method of its manufacture or the uses to which it is put ; for ex- 
ample, plate, crown, flint, and window glass. 

* Wagner, Chemical Technology, 608. 

t Ost. Lehrbuch d. technischen Chemie, 5* Auf., 237. 


Plate glass is cast on a large iron plate or " casting table," made 
up of thick, narrow segments of cast iron, bolted together and planed 
on top. These tables were formerly cast in one piece, and, being 
large and thick, were very expensive. But when put to use, they soon 
became warped and dished, owing to unequal expansion of the top 
and bottom; this caused much loss of time and glass in the subse- 
quent grinding of the plate. The built-up table is much cheaper and 
retains its even surface much longer. 

The melted glass is poured on the table and spreads out in an 
even layer. But to give the plate a uniform thickness and to smooth 
down any inequalities of the surface, a heavy iron roller, travelling on 
adjustable guides at the edge of the table, is passed over it. The 
height of these guides determines the thickness of the plate. Both 
the casting table and the roller are heated before use, so that the glass 
may not be cooled too rapidly. As soon as the plate is rolled, it is 
transferred to the floor of the annealing furnace, or " lehr," which is 
directly in front of the casting table, and which has been heated to the 
temperature of the glass. The floor of the oven consists of a series of 
iron plates supported on rollers ; as each new sheet of glass is intro- 
duced the entire series is moved forward towards the outlet. As the 
furnace is heated only at the inlet end, the temperature gradually de- 
creases towards the outlet, and the glass slowly cools during a number 
of hours. All glass must be annealed. This process probably allows 
the molecules to arrange themselves so that there is no considerable 
internal stress when the mass is cold. Unannealed glass which has 
been suddenly cooled is always under high internal strain, which 
makes it exceedingly brittle, and may even cause it to fly to pieces 
spontaneously, or when slightly scratched. 

When removed from the annealing furnace, the plate is uneven 
and rough, and may be somewhat devitrified on the surface. It is 
fastened on a horizontal table, and heavy cast-iron rubbers are made 
to slide over its surface with a rotary motion, while coarse sand and 
water are sprinkled on it. When the glass is smoothed and of a uni- 
form thickness, it is polished by rubbing with buffers, covered with 
leather or felt, and used with fine emery dust or putty powder. About 
one-half of the thickness of the plate is cut away during the grinding 
and polishing. 

Plate glass is usually a soda-lime glass. The batch is melted and 
refined as has been described, great care being taken to remove all 
the " gall/' which is skimmed off immediately before the casting. An 
especially strong pot is used, which will stand the strain of lifting 

GLASS 205 

from the furnace while full of melted glass. The furnace is constructed 
with brick-lined, cast-iron doors, which open to permit the removal of 
the pot. The melting and annealing furnaces are often joined, so 
that the latter may be heated with waste heat. 

The chief uses of plate glass are for windows and mirrors. A con- 
siderable quantity of " rough plate," unground, as it comes from the 
annealing furnace, is used for skylights and for flooring. 

Window glass is always blown. It is usually a soda-lime glass, 
and the batch is melted and refined in the usual manner, either in 
pots or in tanks. After the refining, the glass is allowed to become 
pasty, and then the blower begins his work. His chief tool is the 
" pipe," a straight piece of iron tubing, four or five feet long, usually 
provided with a mouthpiece. He dips the pipe into, the soft glass, 
which is called " metal," and gathers a lump on the end. Then, by 
blowing through the pipe, while whirling it between the palms of his 
hands, he forms a hollow globe of glass. This is reheated in a special 
furnace ("glory-hole") until soft, rolled on a flat surface, and then 
swung in a vertical circle, with occasional blowing through the pipe, 
until the globe has elongated into a hollow cylinder, closed at one 
end and opening into the pipe at the other. In order to have 
plenty of room for the vertical swinging, the workman stands on a 
bridge placed across a rather deep pit. The closed end of the cylinder 
is reheated until soft, and then. blown out; the small opening thus 
made is enlarged by means of the " widening tongs." The pipe is 
detached by touching its point of attachment with a wet stick, and 
the edges of the still soft glass are trimmed with shears. A hollow 
cylinder, open at both ends, is thus formed, and is cut lengthwise 
with a diamond. It is then put into the flattening furnace, in such a 
position that the cut is on the upper side. The heat being sufficient 
to soften the glass, the cylinder slowly opens, and spreads out on the 
floor of the furnace in a flat sheet. It is then transferred to the an- 
nealing furnace for blown ware. This consists of a long oven, heated 
at one end and cool at the other. A system of endless iron bands 
carries the glass slowly from the hot to the cool end of the oven. 
Sometimes the glass to be annealed is placed on a large horizontal 
table, usually built of slabs of stone, and carefully balanced, so as to 
revolve easily and slowly, by means of a gear, while a segment passes 
through a narrow opening in the side of the flattening furnace, where 
it is exposed to the high temperature. The glass is thus slowly carried 
out of the furnace into a cooler compartment, from which it is re- 
moved when nearly cold. This table is chiefly used for window glass. 


The glass sheets are now cut to marketable size, without any polish- 
ing. As the surface of blown glass is fused and not polished, it is 
brilliant and hard ; hence, it is less easily scratched or etched, and is 
more durable than plate glass when exposed to the weather. 

Glass-blowing is an exceedingly fatiguing labor, and only men of 
strong constitution and good lung power can do it. The mass of 
glass which a good workman will handle at one time averages about 
18 pounds, and from it he will form a cylinder over a yard long and a 
foot in diameter. A skilful glass-blower can form all kinds of glass 
utensils by the use of his pipe and other tools. Wine glasses, tumblers, 
bottles, and lamp chimneys, for example, are often entirely blown. 
But much glass ware is blown in moulds, or pressed. Glass-blowing 
machines are now in use for common bottles, lamp chimneys, fruit 
jars, etc., and in some cases window glass, carboys, and other large 
articles are machine blown. 

Crown glass is a form of blown glass in which the globular balloon 
first blown is flattened by pressing against a flat surface. The end 
of an iron rod is smeared with a coating of melted glass and attached 
to the centre of the flattened surface. The pipe is then detached, 
leaving a small hole. By reheating and rotating the rod swiftly 
about its longitudinal axis, the balloon opens out, forming a circular 
plate or- disk, 4 or 5 feet in diameter with the rod at the centre. But 
they are not of the same thickness at the edge and middle ; where the 
rod was attached, there is a thick rounded mass called the " bull's- 
eye." This must be cut out, 'so large window panes cannot be made 
from crown glass. Thus it is not an economical form of glass-blowing, 
and the industry is practically abandoned. A little is now made to 
supply a small demand for the " bull's-eyes " for decorative purposes. 
Crown glass has a very brilliant surface. 

Flint glass now means any transparent, colorless glass. 

In cut-glass ware, the design is cut in the solid glass, which has 
been given its general form by blowing or pressing. Sometimes the 
design is formed in the pressed ware, and the surface only is cut and 
polished. Glass-cutting is done on a soft steel, copper, or sandstone 
wheel, the cutting edge of which is fed with sand or emery and water. 
The polishing is done on similar wheels of wood, fed with rouge or 
putty powder. Lead glass is nearly always used for cutting, since it is 
softer and more brilliant than other varieties. 

Pressed glass is made by the use of a die or mould ; these moulds 
are expensive, but owing to the great mumber of pieces of the same 
form and design that are made with slight labor, pressed ware is cheap. 

GLASS 207 

" Tough " or " tempered " glass is produced by a special method 
of annealing, the articles so treated being capable of withstanding 
blows and sudden changes of temperature. This tempering is done 
by plunging the article, while still so hot as to be somewhat soft, into a 
bath of oil heated to 100-300 C . This sudden " quenching " hardens 
the surface of the glass, but causes internal stresses. If scratched or 
cut slightly, toughened glass is apt to fly to pieces, sometimes with great 
violence. And even after standing a long time spontaneous fracture 
often occurs. It is mainly used for lamp chimneys. 

A process for making hardened glass plates and window lights is 
employed in which cold metallic surfaces are applied to the glass 
plates while the latter are still plastic. The sudden chilling imparts 
an exceedingly hard surface to the glass, so that it can be used in 
exposed situations, such as in street lamps. 

A compound glass is a recent invention to replace the hardened 
or tempered glass. Articles are formed of two layers of glass, the 
inner layer having a low coefficient of expansion while the outside 
layer has a high coefficient. This glass is particularly recommended 
for lamp chimneys and chemical vessels which must endure sudden 
changes of temperature. The ratio between the two coefficients must 
be very carefully maintained. 

Colored glasses are produced by adding to the ordinary batch cer- 
tain metallic oxides or salts, or even finely pulverized metal. These 
dissolve in the glass, and impart a characteristic color. 

Green glass is produced by the use of ferrous oxide, chromic 
oxide, or a mixture of cupric and iron or chromium oxides. The 
color produced by ferrous oxide is a dull green, of no particular 
beauty. Copperas or iron filings are added to the batch to form 
the ferrous silicate necessary for the color. Chromic oxide (Cr 2 O 3 ) 
imparts a better green. It is usually produced by adding potassium 
bichromate (K^C^Oy) to the batch. If an excess of chromium oxide 
is present, the uncombined portion separates as minute crystals, dis- 
seminated through the glass, producing chrome aventurine. A mix- 
ture of cupric and ferric oxides produces green glass, owing to the 
combined effect produced by these oxides individually. 

Yellow or amber glass is made by adding sulphur or carbonaceous 
matter to the batch, producing sodium or potassium sulphides, which 
color the glass. A common method is to introduce wood or charred 
horn into the melted glass. Cadmium sulphide is sometimes employed. 
No sulphur compound can be used with lead glass. A rich yellow 
stain is obtained by the use of metallic silver or silver chloride ; this 


is much used in making church windows, and was known in the Middle 
Ages. A peculiar greenish yellow, fluorescent glass is produced with 
uranium oxide, but it is expensive. 

Orange glass of various shades is made by adding selenium (as a 
selenate with a reducing agent), or a mixture of ferric oxide and man- 
ganese dioxide. 

Blue glass is made with cobaltic oxide (0203) or cupric oxide. A 
very small percentage (0.1) of cobaltic oxide produces a deep blue 
color. If more than 5 per cent is used, the color is so deep that the 
glass may be ground for pigment (smalt). Owing to the intensity of 
its color, cobalt glass is much used for " flashing " on the surface of 
colorless glass. To do this, the blower dips his pipe into the pot of 
colored glass, and, collecting a small lump, dips it into the pot of color- 
less glass, or vice versa. By blowing he forms a sheet of colorless 
glass which is coated on one side with a very thin layer of colored 
glass, both firmly welded together. Both glasses must have the 
same composition and the same coefficient of expansion. A light 
greenish blue is obtained by the use of a small quantity of cupric 

Violet is produced by a small amount of pyrolusite, free from iron. 
An excess of manganese, especially if much iron is present, gives a 
deep yellow or brown. 

Red glass is made with metallic gold or copper, cuprous oxide, or 
selenium oxide. Gold yields a bluish red, while copper and selenium 
give deep ruby red. For gold ruby a minute quantity of gold chloride 
is added to the batch; on cooling the glass is colorless or reddish 
yellow. The ruby color appears on reheating and is due to colloidal 
precipitation of metallic gold in the glass ; if overheated the color 
changes to a dull red brown, owing to coagulation of the colloidal gold. 
Gold ruby is generally "flashed," owing to its intense color. 

For copper ruby, cuprous oxide (Gu 2 O) as "hammer scale," is used ; 
a small quantity of iron filings may be added to reduce any cupric 
oxide. The pot metal is also nearly colorless or pale green and re- 
quires careful reheating to develop the colloidal copper precipitate. 
This glass is also used for "flashing." 

Selenium oxide yields a beautiful deep red in the pot metal, thus 
avoiding the troublesome second heating ; but the color is less intense 
than with gold or copper, and flashing is not employed. 

White, " opal," or " milk " glass is made by adding cryolite or 
fluorite, with felspar, to the batch for common glass. Calcium phos- 
phate, as bone-ash, may also be used. These substances crystallize 

GLASS 209 

in the glass when the melt is kept near its fusion point for some time, 
and thus cause the opalescence. Large quantities of tin or zinc 
oxides produce a translucent milk glass. 

Black glass is obtained by using a large excess of pyrolusite, iron, 
or copper oxides. The so-called "smoked glass," used for optical 
purposes, contains some nickel. 

Enamel is an easily fusible glass, usually containing lead and 
boric acid, or phosphate or stannate of sodium or potassium. It is 
usually white, blue, or gray, the color being produced by adding 
proper oxides. It is used for coating metallic (iron) vessels, pottery, 
(tiles, flower-pots, bricks, etc.), and porcelain. For cooking vessels 
it must be free from lead, and is composed of sand, borax, soda, and 
calcium phosphate or white clay (kaolin). Enamel must have a co- 
efficient of expansion about equal to that of the iron on which it is 
placed, otherwise the glaze is soon destroyed by heating and cooling. 

Iridescent glass is made by exposing the hot glass to the vapors 
of stannic chloride (SnCU), or hydrochloric acid; these attack the 
surface of the glass and alter its composition. It was formerly sup- 
posed that the art of making this glass was invented by the Romans, 
and later was lost. However, the old Roman glass was not originally 
iridescent, but has become so through exposure to dampness and 
carbon dioxide. The surface has been partly decomposed, the alkali 
dissolving, thus producing a thin layer of glass having a different com- 
position and physical structure from the main body. This thin film 
causes interference of the light rays and produces a play of colors 
when viewed in different positions. 

Mirrors were formerly coated with an amalgam of tin. Tin foil 
was covered with mercury, and the glass, carefully cleaned, was laid 
on the amalgam, excess of mercury being forced out at the sides, and 
the amalgam adhering firmly to the glass. 

But the silver mirror is now the only kind made. A coating of 
metallic silver is deposited on the glass from an ammoniacal solution 
of silver nitrate by the use of a reducing agent. Ammonium tartrate, 
or a solution of glucose or milk sugar in caustic soda, is generally 
used for this purpose ; or aldehyde is sometimes used. The glass is 
carefully cleaned and covered with the silver solution containing the 
reducing substance, ^nd heated gently on a steam or hot-air bath. 
The thin layer of metallic silver deposited adheres to the glass, and 
is washed and dried, and covered with a protecting varnish to pre- 
vent the hydrogen sulphide in the air from tarnishing the reflecting 


Plate glass is generally used for the best mirrors. Blown glass, 
which is used for the cheaper ones, is very apt to contain bubbles 
and strise, causing distortion of the image. 

Tradition assigns the discovery of glass to the Phoenicians. Glass 
making is a very old industry, and was known to the early Egyp- 
tians, since glass beads have been found in mummy cases at least 3000 
years old. Glass articles have also been found in the excavations at 
Nineveh. From Egypt, the industry was transferred to Rome, and 
on the fall of the Western Empire the art was carried to Byzantium. 
Byzantine glass attained a high degree of perfection ; but in the middle 
of the thirteenth century Venice became the centre of the industry, 
and Venetian glass-blowers were remarkably expert in the production 
of beautiful and delicate patterns. Finally, Bohemia took the lead 
in the manufacture of glass, and has retained a front rank ever 

Window glass was made by the Romans to a small extent, and 
specimens of such glass were taken from the ruins of Pompeii. In 
England, it first came into use in houses during the reign of Elizabeth, 
but previously to this it had been used in cathedrals and churches. 
From the records of York cathedral, it is shown that during the time 
of Archbishop Wilfrid (669-709 A.D.) " glass was placed in the win- 
dows so that birds could no longer fly in and out and defile the sanc- 
tuary." The contract for the glass in the great West Window, given 
by Archbishop Melton, is dated 1330. The work was finished before 
1350, and the price paid was 6d. per square foot for white, and Is. 
per square foot for colored glass. This window is 54 feet high by 30 
wide, and is to-day regarded as one of the finest examples of stained 
glass in England. The great East Window (77 feet high and 32 feet 
wide) was glazed by John Thornton in 1405-1408, for which he re- 
ceived 4s. per week. These examples demonstrate the high degree 
of perfection to which the glass industry had advanced during the 
Middle Ages. 

At the present time, Belgium and England lead in the production 
of window and plate glass, while Germany, France, and the United 
States also manufacture enormous quantities. Austria and Germany 
are the leading producers of blown ware. 

GLASS 211 


Glass Making. Powell, Chance and Harris. 1883. 

U. S. Census, 1880. Report on the Manufacture of Glass. J. D. Weeks. 
Die Glas-Fabrikation. R. Gerner, Vienna, 1880. (Hartleben.) 
Handbuch der Glas-Fabrikation. Dr. E. Tscheuschner, Weimar, 1885. 

Die Fabrikation und Raffinirung des Glases. Wilhelm Mertens, Vienna, 

1889. (Hartleben.) 
Verre et Verrerie. L. Appert et Jules Henri vaux, Paris, 1894. (Gau- 

thiers-Villars et Fils.) 
Journal of the Franklin Institute. 1887. Glass Making. C. H. 

Proceedings of Engineers' Society of Western Pennsylvania. 1895, 119. 

A Study of Glass. Robert Linton. 

Elements of Glass and Glass Making. B. F. Riser. 1899. 
Jena Glass. H. Hovestadt. 1902. 
Journal of the American Chemical Society. 1902, 893. G. E. Barton. 


Clays are natural hydrated silicates of aluminum, formed by the 
weathering of felspar or felspathic rocks, such as granite. The hy- 
drolysis of felspar may be represented thus : 

A1 2 O 3 , K 2 O, 6 SiO 2 + CO 2 + 2 H 2 O = 

A1 2 O 3 , 2 SiO 2 2 H 2 O + K 2 CO 3 + 4 SiO 2 . 

Clays which have not been transported by natural waters from 
the place where they were formed are called primary; secondary clays 
have been washed from their original beds and deposited elsewhere. 
Primary clay derived from pure felspar contains little impurity other 
than silica, and is called kaolin, or China clay. It is a white, powdery 
mass, essentially hydrated silicate of aluminum and silica, nearly all 
the alkali having been leached out. Kaolin is characterized by its 
capacity to undergo progressive hydration in contact with water ; in 
the early stages this hydration is merely on the surface of the particles, 
but as it penetrates deeper and becomes more complete, the clay as- 
sumes markedly colloidal properties. Its aqueous suspensions become 
tolerably permanent (e.g. river silt), and in the solid state it is very 
plastic, i.e. it can be readily moulded, and surfaces pressed together 
coalesce. The water of hydration can be driven off by heat, where- 
upon the plasticity is lost, and the mass becomes hard and stonelike. 
These facts are the basis of the ceramic industry. 

The rocks from which primary clays are formed are seldom homo- 
geneous, and the various components have different rates of decom- 
position and hydration. Tne clay then consists of a mixture of 
particles of different composition and colloidal character. The 
formation of secondary clays usually involves mixing of various pri- 
mary deposits, and the particles of such clays are more diverse in 
character. Kaolin itself is almost infusible, but if the clay contains 
in an appreciable amount particles of low melting point (unweathered 
felspar, quartz, and metallic oxides), on heating, these finely divided 
particles disseminated in the mass serve as a flux for the rest of the 
clay, and cause it to soften, or even to melt, at temperatures far below 
the fusion point of the major part of the mass. Thus a clay may be 
of such composition that if rendered homogeneous (e.g. by melting) 
its fusion point is very low, but in the clay itself none of the various 



particles are of low fusion point ; such a clay will soften only at very 
high temperatures, but once softening begins, the mass will quickly 

Very difficultly fusible clays are generally primary, as among 
the constituent particles of secondary clays, some are likely to be of 
low melting point. Fire-clays are very difficultly fusible. They are 
usually found underlying coal beds. In composition they are kao- 
lins containing free silica as quartz. They may contain a little more 
iron than good China clay, but are free from alkalies. 

Secondary clays have had more opportunity for thorough hydra- 
tion than primary, and are hence more plastic ; these colloidal clays 
are called pipe- or ball-clays, and are also known as " fat " clays to 
distinguish them from the non-plastic or " lean " clays. 

Fat clays absorb much water and have great binding power, so 
that they are easily shaped by the potter. But on drying, and es- 
pecially when burned, they shrink much. This shrinkage is coun* 
teracted by mixing with the fat clay a certain amount of "leaning" 
material, such as silica, pulverized burned clay, or " grog," the ground, 
unglazed body of pottery. Fat clays are usually more fusible than 
lean clays. 

Highly hydrated, plastic clays can lose a large part of their water 
without much modification of structure, if dried at temperatures 
below 100 C. They still rehydrate if wet. All clays lose water 
of chemical composition below a red heat, and then lose capacity for 
rehydration and plasticity. All stages of dehydration are accom- 
panied by shrinkage, roughly proportional to the original plasticity ; 
if excessive, this causes cracks in drying and burning ; it is controlled 
by mixing plastic with non-plastic clays, or " grog." Since the physi- 
cal properties of the product are determined by the proportions of 
the various clays employed, in making ware to* meet definite require- 
ments, a complex mixture of clays is usually necessary. 

Clay is used in all degrees of hydration ; in the form of a colloidal 
suspension, or " slip," it makes possible the forming of articles by 
" casting " in plaster moulds ; when dehydrated to the form of a 
plastic solid mass, it is used on the potter's wheel. Dried at a low 
temperature, it loses its plasticity and becomes hard, as in sun- 
dried bricks (adobe) ; it loses chemically combined water below a 
red heat, and an association of the aggregates occurs, which renders 
it inert and incapable of rehydration; at higher temperatures, the 
particles of lowest melting point fuse and cement together the remain- 
ing granules ; finally fusion may go so far that the mass becomes semi- 



vitreous, but complete fusion is avoided to prevent deformation of 

Because of their nature and origin, clays are complicated mix- 
tures, each component differing in chemical behavior for varying 
degrees of hydration; hence neither the empirical analysis, nor the 
so-called " rational analysis " is of much value to the potter. The 
latter is a determination of "clay substance" or kaolin, felspar, 
quartz, etc. (by their different behavior to solvents) and merely 
emphasizes the differences between clays. The plasticity, fusibility, 
shrinkage, and color of the product after burning, and its coefficient 
of expansion are the important properties, and these must be experi- 
mentally determined for each clay. The variations in clays are 
shown in the following analyses : 











SiO 2 .... 









A1 2 3 

CaO . . . . . 


Alkalies .... 
H 2 ...... 







Clay substance 






Quartz .... 






Felspar .... 





' 25.31 






All clays have a peculiar and characteristic odor when breathed 
upon or wet. 

The preparation of clay for the potter is simple. It is mined, 
and allowed to weather for several months, which increases its plas- 
ticity by promoting hydration. This is probably aided by certain 

* Lehrbuch der technischen Chemie. 5 te - Auf. H. Ost, 262. 
t Chemistry of Pottery. K. Langenbeck, 10, 111, 165. 


enzymes, of bacterial origin in some cases. Fine clays to be used for 
the better grades of ware are then thoroughly " slipped " with water 
in a " blunger " (a vat with mechanical stirrers), and thus levigated. 
The coarse particles of quartz, mica, and undecomposed felspar are 
separated, and only the clay substance, with a little finely divided 
quartz, remains in suspension. The fine mud, called " slip/' obtained 
by settling the wash waters, is pressed in cloth bags or it is filter- 
pressed. It is then ready for use. 

Ceramics comprise two general divisions : (a) articles having a 
non-porous body, and (6) articles having a porous body. Non-porous 
ware is hard, impervious to liquids and gases, and has a semi-vitrified 
appearance on the fractured surface. It is burned at a very high 
temperature, and is chiefly made from kaolin, with just enough 
plastic material to enable the workman to form the desired article. 
This division includes porcelain and stoneware. Porous ware is less 
dense, has an earthy appearance on the fractured surface, and per- 
mits the passage of gases and liquids through its pores. It is made 
from plastic clays, and burned at a low or moderate temperature. 
It comprises bricks, terra cotta, common crockery, and some kinds 
of stoneware. 

There are two kinds of porcelain, the hard and the soft, or 
" fritted." Both are harder than glass, and very resistive to chem- 
ical action. 

Hard porcelain softens only at the highest attainable tempera- 
ture, and, when burned, forms a perfectly homogeneous mass, which 
is translucent. The body is composed of kaolin, quartz, and felspar, 
in definite proportions. It is glazed with pure felspar, or a mixture 
of quartz and felspar, with sufficient lime to form a difficultly fusible 
glass. This glaze, which must have the same coefficient of expan- 
sion as the body, is very perfectly welded on to the body, by a second 
burning at a high temperature ; and no distinct line of demarcation 
between the body and the glaze can be seen on a fractured surface. 
Berlin, Sevres and Meissen ware are examples. 

Soft porcelain consists of a kaolin body, with ball-clay, bone-ash, 
and felspathic materials added. This is burned at a high tempera- 
ture, and glazed with a lead-boric-acid glass, which is fused on to its 
surface by a second much lower heating. The glaze does not pene- 
trate so perfectly, but forms a more superficial layer than is the case 
with hard porcelain. English china, stone ware, and Parian ware are 
soft porcelain. 

In preparing the clay for porcelains, the powdered materials are 


thoroughly mixed, wet, and the " slip " kneaded and allowed to age 
for several months. The articles are formed on the potter's wheel, 
a horizontal revolving table, driven by foot or machine power. Some- 
times the slip is cast in porous moulds of gypsum or burned clay, 
which absorb the water, leaving the mud on the face of the mould. 
Or the partly dried mud is pressed in moulds to form one surface 
of the article, the other being completed on the wheel, as is the case 
with dishes and plates. The articles are very slowly dried at atmos- 
pheric temperature, and then burned at a low red heat, to give them 
sufficient coherence to permit of glazing. 

The finely powdered glaze mixture is sthred up with water to 
form a cream, into which the articles are dipped and at once with- 
drawn. A layer of the glaze adheres to the surface, and, after dry- 
ing, the article is ready for the second or glaze burning. In order 
to protect them from direct contact with the fire in the kiln, they 
are enclosed in fire-clay boxes, called " saggers." These are piled 
in the kiln in columns or " bungs," extending from the bottom to 
the top. In order to allow sufficient freedom for shrinkage, the 
porcelain is supported on a " cockspur," a small tripod of fire-clay. 
The contraction of porcelain on burning is nearly 13 per cent of its 
original volume. After burning, the ware is sorted ; much is lost 
owing to warping, to bubbles in the glaze, and to discolorations result- 
ing from smoke and from iron oxide in the material. 

The body of all ware to be glazed is called " biscuit " after the 
first firing ; that of soft porcelain which has been hard fired is called 
" Parian." Both are used for statuettes, medallions, and reliefs. 

Stoneware, which is also a non-porous body, is made from refrac- 
tory material, and burned at high temperatures. But the color 
of the resulting ware may range from white and gray to yellow and 
brown. It is not attacked by chemicals, and withstands tempera- 
ture changes fairly well. The finest quality is the well-known 
" Wedgwood " ware, which comes in various colors, and is usually 
not glazed. The gray stoneware, decorated with blue, so much 
used for drinking-mugs and ornamental vases, is also of this group. 
Yellow and brown varieties are much used for mineral water-bottles, 
bombonnes, condenser tubes, and glazed pipes in chemical factories. 
The clays are less pure than those for porcelain, and the ware is burned 
without saggers, at a very high temperature. A " salt glaze " is 
used, to form which common salt is thrown into the kiln, and, - vol- 
atilizing, combines with the silicates of the stoneware to form double 
silicates of soda and alumina on the surface of the ware ; or the 


articles are " slip glazed " by applying an easily fusible clay as slip/* 
before firing. 

The kilns for potters' use are of several kinds. A common form 
is the up-draught kiln, in which the flame enters at the bottom and 
passes up between the " bungs," and out at the chimney above. 
A better type is the down-draught kiln, which is sometimes built 
in two stories. The lower story is filled with the ware to be fired 
at the highest temperature, and the upper with that to be burned 
at a less heat. The flame from the grate passes up through flues 
in the kiln walls, and enters the lower chamber near the top. It 
then goes down between the bungs, and, through openings in the 
floor, into other flues in the walls, around the upper chamber, and 
thence to the chimney. This kiln is economical of fuel, affords 
very even temperature in the lower chamber, and utilizes the heat 
which is lost in the up-draught kiln. A special form of Hoffmann's 
ring furnace (p. 185) is also employed for pottery and brick burning. 
In a new form of kiln, the bungs are arranged on cars, which travel 
slowly through a long gallery, towards the firing chamber. The 
waste heat from the hot chamber enters the gallery at the end next 
the firing room, and, coming in contact with the pottery, heats it to 
a temperature corresponding to its distance from the inlet flue. The 
cars move through the furnace into a second long gallery, where the 
heat from the saggers warms the air which is passing into the fur- 
nace, thus utilizing the waste heat. The firing compartment usually 
contains two loaded cars ; and the grate being at one end of the fir- 
ing room, the pottery in each car gets a preliminary firing before it 
reaches the hottest part of the kiln. As soon as one car is fired, it is 
pushed into the cooling gallery, the rear car is moved into the hottest 
compartment of the kiln, and another is introduced from the pre- 
liminary warming gallery. This furnace is economical of fuel, gives 
an even temperature, and the time of firing being greatly reduced, 
there is less loss of saggers and pottery. 

Porous ware, the second division of ceramics, is manufactured 
extensively in all countries. The finest grade is " faience." This is 
made from a white clay, which is washed, levigated, and aged, much 
as for porcelain. The better grades are burned in saggers at a high 
temperature, and glazed with a transparent lead glaze at a much 
lower heat. Majolica also belongs to this group, being a colored porous 
body, covered with a non-transparent glaze. 

Between faience and common pottery no sharp line can be drawn. 


The color ranges through cream, yellow, brown, and red, and the 
body consists of more or less fusible clay, with a still more fusible 
lead glaze, which is often colored with metallic oxides. The clays 
for common pottery are generally " slipped," and strained through 
fine sieves to remove stones and coarse grains. The articles are 
fashioned on the potter's wheel and are air dried. They are then 
dipped in a glaze made of litharge and clay, shaken to a cream with 
water. Or the dry mixture is powdered over the surface from a pepper 
box. Or they are given a " salt " glaze as before described. They 
are burned without saggers, and at a temperature only sufficient to 
fuse the glaze. 

Tiles are a special form of pottery, consisting of flat, thin plates, 
much used for floors, panels, and architectural purposes. They are 
finer ware than common brick, and more care is taken in the prepa- 
ration of the body and in the burning. There are three classes, 
vitrified, encaustic, and glazed. 

Vitrified tiles consist of single pieces, made by one burning at a 
very high temperature, so that the entire body of the tile is semi- 
fused. They are not glazed, and are a form of stoneware much 
used for pavements and floors, because of their hardness. 

Encaustic tiles are made from two or more clays, generally of 
different colors. A facing of fine clay may be put on a back of com- 
moner quality. The ornamental design is made by inlaying the 
face with other clays, which burn to different colors. All the mate- 
rials must have the same coefficient of expansion, so that no cracks 
form between the different parts of the design. These tiles are 
generally used for ornamental purposes, and are often covered with 
a transparent glaze, necessitating two burnings. 

Glazed tiles are made with a body (which may consist of more 
than one clay) of uniform color, covered with a transparent glaze, 
colored or not, according to the effect desired. 

The dry clay, flint, felspar, Cornish stone,* " grog," and other 
materials in the mixture for the body of the tile, are put into a re- 
volving drum (Alsing mill), along with a number of round flint stones. 
After five or six hours' grinding, the mixture is complete. The dry 
powder is then sifted through a fine sieve. There are two methods 
of forming the tile, the " dust body " and the " wet body " process. 
In the dust body method the sifted clay mixture is dampened by spread- 

* Cornish stone is partly weathered felspar, being thus a mixture of kaolin, 
felspar, quartz, and mica. It is mined in England, and much used as a flux and 
fusible ingredient in porcelain and tiles. 


ing on a wet plaster of Paris floor. It is shovelled over and allowed 
to remain on the floor until the particles of clay will just stick to- 
gether when pressed in the hand. It is then filled into a metallic 
mould which contains the intaglio for relief designs ; it is then heavily 
pressed in a screw or hydraulic press. This compacts the clay, 
and gives sufficient coherence, so that the green tile may be removed. 
It is exceedingly brittle, and must be handled very carefully. It 
is well dried in a room where there is a good circulation of air. To 
prevent discoloration, tiles are burned in saggers in which they are 
loosely packed in quartz sand to prevent their twisting and bending, 
since they become very soft at high temperatures. 

In the wet body process the slip is moulded in plaster of Paris 
moulds. After standing half an hour, or more, until the water has 
all been absorbed by the plaster, the clay cast is removed, dried 
slowly, and burned as in the case of dust body tiles. 

Glazes, both for hollow ware of all sorts and for tiles, are of three 
kinds, engobe, enamel, and transparent. 

The engobe is a fusible clay, felspar, or alkali, applied in a very 
thin coating. It forms a thin glaze, usually opaque, which is intended 
to support a second thicker glaze or enamel. 

Enamels are usually transparent glazes, holding in suspension 
opaque substances such as oxide of tin. A mixture of litharge and 
tin oxide (" ashes of tin ") is very often used for enamel. 

Transparent glazes are practically lead or lime glass. This is 
sometimes, though rarely, used as " raw glaze," i.e. the materials are 
ground fine, mixed, and applied to the ware as a cream with water. 
This is difficult to do, owing to the great density of the litharge, 
which settles out of the cream, on standing even a short time. To 
avoid this separation and loss, and to allow the use of substances 
soluble in water, e.g. borax, soda-ash, or boric acid, the glaze is gen- 
erally " fritted " or semi-fused, before making it into a cream with 
water. The powdered and thoroughly mixed material, together 
with coloring substances if desired, is heated in a sagger until it 
forms a coherent mass, but is not completely fused. The frit is then 
powdered in a ball mill. Fritted glaze is much more uniform than 
raw, and there is no tendency to segregation of its components. 

In all kinds of glazed ware, it is very essential that the glaze 
and body shall have the same coefficient of expansion, or cracking 
of the glaze is liable to occur. This is called " crazing," and is 
caused by the glaze contracting too much in cooling ; the scaling off 


of glaze and attached body from high points of the tile, called " shiv- 
ering," is caused by insufficient contraction of the glaze. To pre- 
vent these defects the glaze or the body is so modified that the co- 
efficients of expansion are the same. The exact adjustment of this 
factor is a matter of experience. The usual methods employed 
are, to render the body less plastic by the addition of lean clay, 
grog, or quartz, thus increasing the silica, which increases the expan- 
sion of the body ; or to modify the glaze by the addition of silica 
or boric acid for greater expansion, or of lime, lead, or alkali, to 
increase the contraction. Boric acid, lead, and alkali make it more 
fusible, and the temperature of the intended burning must be kept 
in mind when adding these ingredients. Boric acid and lead also 
increase the brilliancy of the glaze. The addition of certain color- 
ing matter to glazes also increases the tendency to craze. Alumina 
is essential in a glaze to prevent devitrification during the burning. 

Terra cotta has a soft porous body, and is not glazed. Its color 
depends on the character of the clay. Generally a highly ferruginous 
clay is used, which has a deep red color when burned. It is exten- 
sively employed for architectural effects and for tiles. 

Bricks are probably the most important of the porous ware. 
They are made from common clay, which usually contains consider- 
able impurity, lime and iron oxide often being present in large quan- 
tities. The preparation of the clay is a simpler process than for 
pottery. After digging it is usually weathered for several months, 
and then screened, to remove pebbles of quartz or limestone.* 
It is then " pugged " or " tempered," by mixing thoroughly with 
water and tlie ingredients to make the desired " body " ; in the 
case of a fat clay, these are sand, grog, or other clays. This is done 
in a "pug mill," a tank containing a revolving stirring apparatus, 
which pushes the mass out at the bottom in proper condition to be 
used at once. The paste is moulded into bricks, by hand for the finer 
sort, and by machinery for the common grades. The latter are apt to 
be uneven and rough. The moulded bricks are dried in the air, usually 
in the yard, under a light shed. They are turned over frequently 
during the drying, which must not be too rapid, lest the bricks crack. 

The firing is done in kilns which may be built of the air-dried 
brick, numerous channels being left for the passage of flame and hot 
gases. This mode of burning results in several grades of brick, 

* Limestone pebbles are very injurious, since the burning converts them into 
lumps of lime within the brick, and when the latter is wet or exposed to weather the 
lime is hydrated, and, expanding, disintegrates the brick. 


owing to the unequal distribution of heat. Or closed kilns, such as 
the Hoffmann ring furnace (p. 185), may be used. This gives a more 
even product than the open kiln. 

In this country wood and coal are used for fuel, but gas is fre- 
quently employed abroad. The temperature in the kiln for common 
brick seldom goes higher than 1000 C. ; but for hard, paving brick 
it may be raised to 1200 or 1300 C., producing incipient fusing. 
The heat also affects the color of some bricks ; high temperature 
yields a dark red, gray, or bluish black, according to the amount of 
ferroso-ferric oxide (Fe 3 O 4 ) formed. Clays containing much lime 
yield yellow or cream-colored brick, if iron is also present. 

Common brick will fuse if exposed to high heat, and are not suit- 
able for lining fireplaces, furnaces, or ovens. 

Fire-brick are made from fire-clay, with the addition of a large 
amount of " grog " or silica. These must resist great heat, and not 
shrink nor warp. The clay is prepared similarly to that for common 
brick, but more care is taken in the mixing. The bricks are also 
heavily pressed to increase the density. The burning is at the highest 
temperature possible, so that no shrinkage will occur later when the 
bricks are in use. They are brittle, and must be supported by a 
backing of common hard brick. 

Bricks, the body of which are magnesia, chromite, silica, etc., 
mixed with just enough plastic clay to make them workable and to 
cement the grains of the body when burned, are largely employed in 
chemical furnaces to withstand high temperature corrosion of various 
types of charge. 


Handbuch der gesammten Thonwaarenindustrie. Bruno Kerl, Braun- 
schweig, 1879. 

Traite des Arts ceramiques ou des Poteries. Alexandre Brongniart. 

Lecons de ceramique. A. Salvetat. 

La Faience. Th. Deck. 

Report on the clay deposits of Woodbridge, South Amboy, etc. Public 
Documents of New Jersey. 

A Practical Treatise on the Manufacture of Bricks. C. T. Davis. 

Pottery and Porcelain of the United States. E. A. Barber. 

Die Steingut-Fabrikation. Gustav Steinbrecht, Leipzig, 1891. 

Ziegel-Fabrikation der Gegenwart. Herman Zwick, Leipzig, 1894. 

Seger's gesammelte Schriften. H. Hecht und E. Cramer, Berlin, 1896. 

The Chemistry of Pottery. Karl Langenbeck, Easton, Penn., 1895. 

Clays. Heinrich Ries, New York, 1906. (Wiley & Sons.) 

Clay and Pottery Industries. J. W. Mellor, London, 1914. 

The Silicates in Chemistry. W. Asch and D. Asch. Trans, by A. B. 
Searle, New York, 1914. 


Pigments are mineral or organic bodies, usually insoluble in 
water, oils, and other neutral solvents, and are used to impart color 
to a body, either by mechanical adhesion to its surface or by ad- 
mixture with its substance. In most cases there is no chemical 
combination between the pigment and the body it colors. 

The color of a pigment depends upon the amount and kind of 
light which it reflects. It is essential that the pigment be opaque 
(possess large capacity to absorb transmitted light) in order that 
it may have a good " covering power " or " body " ; i.e. it should 
entirely conceal the surface to which it is applied. Many pig- 
ments are prepared by chemical precipitation, but some of the 
most important are not. 

Pigments form the basis of paint, which consists of a mixture of 
a pigment with some drying oil, or with water containing gum or 
size. It is used for decorative and protective purposes ; if used for 
outside work, the pigment should be insoluble in water. 

The durability of a paint depends on the chemical stability of 
the mixture of pigments and vehicle composing the film, and on its 
mechanical strength, resistance, and impermeability. These proper- 
ties are best secured by using a mixture of relatively coarse and fine 
pigments ; the first form a skeleton of large particles, giving strength 
and rigidity, and the latter render the mass impermeable by filling 
the voids between the coarse particles. Thus while coarse pigments 
(barytes, silica, chalk, etc.) are almost valueless alone, their addition, 
in quantities up to 10 per cent, greatly increases the wearing 
qualities of paints containing fine pigments (white lead, zinc oxide, 
etc.). Furthermore mixtures of fine pigments, which differ from 
each other in size of particle, are better than the pure pigments 
alone. This has been established in the case of white pigments ; 
with colored pigments, chemical changes in the pigment, vehicle, or 
both, frequently complicate the matter. 


White lead is the most important of all pigments, and is a 
very old one, the native carbonate, cerussite, having been used by 
the Romans. But as this mineral is restricted in its distribution 



the artificial product was in time brought into use. The so-called 
Dutch process of making white lead is the oldest known, reference 
being made to it as far back as 1622. With a few modifications, it 
is still in use, and yields a product which for many purposes is 
preferred by painters to the lead manufactured by the numerous 
newer processes. It usually has more covering power and a better 

White lead is a basic lead carbonate, and analyses of the best 
samples give as constitutional formula about 2 PbCOs, Pb(OH)2, in 
which there are two molecules of PbCOs to one of hydroxide. This 
appears to be the best ratio. But the white lead of trade varies 
a good deal, according to the method and conditions of making. 
In some cases it is nearly pure PbCOs, and in others the propor- 
tion of carbonate to hydroxide is as high as three to one, or more. 
But some hydroxide is necessary in order that the white lead may 
have a good covering power. Then, too, the hydroxide is sup- 
posed to combine with the oil chemically to form a " lead soap," 
which perhaps dissolves in the 'excess of oil to form a kind of 

There are three general methods employed in white lead making, 
besides numerous patent processes. These are : 

The Dutch, or Stack process. 

The German, or Chamber process. 

The French, or Thenard's process. 

The Dutch process consists in exposing sheet lead to the direct 
action of moisture, acetic acid vapors, and carbon dioxide. The 
corrosion is effected in earthenware pots 8 inches in depth by 5 inches 
in diameter, glazed inside, and made in the form of crucibles, each 
containing a shelf. On this shelf is a spiral or " buckle " of thin 
sheet lead, made by rolling up a sheet of lead 2 feet long by 4 inches 
wide; or cast buckles of various forms to expose a large surface to 
the fumes, may be used. In the lower compartment is dilute acetic 
acid, containing from 3 to 5 per cent C 2 H 4 O 2 . A large number of 
these pots so charged are packed in a shed or brick building, having an 
opening on one side reaching from the ground nearly to the roof. A 
layer of ashes is spread over the floor first, and then a layer 4 or 5 
feet thick of spent tan bark which is moist and ready to ferment. 
This is well packed down, and the pots placed side by side upon it 
until the whole space is filled, excepting about 6 inches next the walls, 
which is solidly filled in with the tan. More lead buckles or lead grat- 
ings are placed across the tops of the pots, so as to form a layer of 


metallic lead about 4 inches deep. Then about 6 inches above this, 
and supported by timbers or blocks, is a board floor upon which the 
next layer of tan, about one foot deep, is placed, and the pots upon it 
as above described. The doorway is boarded up as the filling continues, 
and the " stack," as the alternate layers of pots and tan are called, 
is carried to within a few feet of the top of the shed. For a stack 
20 by 12 by 18 feet in size, 40 or 50 tons of lead are required, about 
3 tons of lead and 200 gallons of acid being used in each layer of pots. 
Very soon after packing an active fermentation of the tan sets in, 
the temperature rising to about 55-60 C. This heat is sufficient 
to vaporize the acetic acid and water, and these vapors attack the 
metallic lead, forming a basic lead acetate. Great quantities of 
carbon dioxide are liberated during the fermentation, and this decom- 
poses the lead acetate, forming basic lead carbonate, or white lead. 
The reactions, aside from those of fermentation, may be represented 
by the following : 

1) Pb + 2 C 2 H 4 O 2 = H 2 + Pb(C 2 H 3 O 2 )2. (Normal lead acetate.) 

2) 3 Pb(C 2 H 3 O 2 ) 2 + 2 H 2 O = 2 Pb(C 2 H 3 O 2 ) 2 , Pb(OH) 2 + 2 C 2 H 4 O 2 . 

3) 2 Pb(C 2 H 3 O 2 ) 2 , Pb(OH) 2 + 2 CO 2 + 2 H 2 O = 

2 PbCO 3 , Pb(OH) 2 + 4 C 2 H 4 2 . 

Some authorities consider the reactions to be as follows : 

a) Pb + H 2 + O = Pb(OH) 2 . 

b) Pb(OH) 2 + 2 C 2 H 4 O 2 = Pb(C 2 H 3 2 ) 2 + 2 H 2 O. 

c) Pb(C 2 H 3 O 2 ) 2 + 2 Pb(OH) 2 = Pb(C 2 H 3 O 2 ) 2 , 2 Pb(OH) 2 . (Basic lead 


d) 3 SPb(C 2 H 3 2 ) 2 , 2 Pb(OH) 2 ( + 4 CO 2 = 

3 Pb(C 2 H 3 O 2 ) 2 + 2 fPb(OH) 2 , 2 PbCO 3 J + 4 H 2 O. 

Thus the acetic acid, or the neutral lead acetate, is regenerated, 
and attacks more of the metallic lead, and the process repeats itself. 
But the action is very slow, the time usually allowed for a stack to 
work being about three months. If horse dung is used instead of 
the tan bark, as it formerly was, the process is quicker, taking about 
two months; but the sulphur compounds formed in this fermenta- 
tion darken the white lead more or less. 

Usually the metallic lead is nearly all corroded and converted 
into white lead, but never completely, and the product is seldom 
equally good in all parts of the stack. Well-corroded buckles retain 


their shape, although they become rather more bulky and of a gray- 
ish white color, and have a firm, porcelain-like structure. When soft 
and powdery, the product is not so satisfactory. The " corrosions " are 
taken to the grinding room and put through rolls, which break them 
into fine powder; the uncorroded lead is rolled out into plates and 
scales, which are retained on the sieves when the mass is screened. 
The white lead passes through, and is reduced to a fine pulp by wet 
grinding in an edgestone or horizontal mill, and then levigated; 
salt is often added to the water to accelerate the sedimentation. 
The coarser particles from the first settling tanks are returned to the 
mills and reground. In the final settling tanks is left a heavy white 
mud, which may be dried in steam-jacketed copper pans. Usually 
the wet mud is mixed with about 9 per cent of raw linseed oil, in me- 
chanical mixers of the helical screw type. The oil combines mechani- 
cally with the pigment and the mixture settles, while the water, now 
almost free from lead, is drawn off from above. The paste of lead in 
' oil contains so little water that it is not further dried as a rule, but 
is packed directly for shipment. This method avoids the dust pro- 
duced in handling the dry pigment. 

The white lead comes in trade either dry or mixed with about 9 
per cent of raw linseed oil. If it is slightly yellow, due to stains from 
the colored liquids in the tan, or to tarry matters in the acetic acid, 
or to overheating in the drying, a little indigo or Prussian blue may 
be added to neutralize this. One ton of the metallic lead yields about 
1 J tons of the white lead ; but the process is always somewhat un- 
certain both as to quantity and quality of the product. Sometimes 
very little corrosion takes place, and this may vary in different parts 
of the stack. The process is slow, a large plant is required, and the 
capital invested lies idle; hence the price of white lead is somewhat 
higher than the simplicity of the method would at first glance appear 
to warrant. 

To obtain good results by the Dutch method the lead must be 
very pure. If any silver, copper, or iron is present, the color of the 
white lead will be damaged. Antimony, arsenic, bismuth, and zinc 
are said to retard the corrosion very much. 

The German, or chamber, process is an artificial method of pro- 
ducing about the same conditions as prevail inside the stack in the 
Dutch method. The reactions are the same. Lead plates are hung 
or arranged on shelves in a closed chamber, provided with a door and 
window for filling and for watching the process. Dishes of acetic 
acid are placed on the floor, or acetic acid vapor is introduced from 



stills outside, the room is heated by steam to about 38 C., and carbon 
dioxide is introduced. This is much more rapid than the Dutch 
method, usually requiring about five weeks, but the quality of the 
product is not so satisfactory. There are difficulties to contend with 
in the rate of flow of the acid vapors, steam, and carbon dioxide, and 
in the regulation of the temperature. Too much acetic acid vapor 
causes loss of lead as neutral acetate ; too much carbon dioxide pre- 
cipitates normal lead carbonate; too little acetic acid or too high 
a temperature, with excess of water vapor, may form lead oxide, 
which, being yellow, injures the product. Many modifications of 
this process have been invented, and some of them are worked more 
or less successfully. In one form, the chamber is fitted with tracks, 
on which cars are run, carrying the sheet lead in frames. A car 
can be run out, and another introduced, without much loss of time 
or cooling of the chamber. The white lead made by this method is 
ground and levigated as already described. 

Carter's process is much used and produces white lead of good 
quality. Atomized lead, made by blowing a jet of superheated steam 
against a fine stream of melted lead, is put, in two-ton charges, into 
rotary wooden cylinders, 6 feet in diameter by 10 feet long; dilute 
acetic acid is sprayed in, and carbon dioxide with air admitted through 
the centre of the head, while the barrel revolves slowly. The con- 
stant stirring accelerates the process, and corrosion is complete in 
about 15 days. The tumbling of the mass also serves to pulverize 
the white lead. The carbon dioxide is made by combustion of good 
coke with excess of air. Coleman's process is similar but the carbon 
dioxide is used under pressure. 

The French process, or ThenarcTs method, depends on precipitation 
of a basic lead carbonate from a solution of a basic salt by means of 
carbon dioxide. The solution generally used is a basic lead acetate, 
prepared by boiling litharge with neutral acetate. The reactions 

1) 2 PbO + Pb(C 2 H 3 O 2 ) 2 + 2 H 2 O = Pb(C 2 H 3 O 2 ) 2 , 2 Pb(OH) 2 . 

2) 3[Pb(C 2 H 3 O 2 ) 2 - 2 Pb(OH) 2 ] + 4 CO 2 = 

3 Pb(C 2 H 3 2 ) 2 + 2 [2 PbCO 3 , Pb(OH) 2 ] + 4 H 2 O. 

The reactions are carried out in the apparatus shown in Fig. 87.* 
The litharge is mixed with the solution of neutral lead acetate in 
the tank (A), which is heated by a steam pipe. When saturated, 
the mixture is run into the settling tank (B), where the undissolved 

* After Hurst, Painter's Colours, Oils and Varnishes. 



litharge deposits. The clear solution of basic lead acetate is then 
run into the precipitating vessel (C), where it is treated with carbon 
dioxide, introduced through the pipes (D, D). The basic lead car- 
bonate falls as a heavy white precipitate, while a solution of neutral 
lead acetate remains. After settling, the solution is drawn into the 
tank (E), from which it is pumped back into (A), where, after add- 
ing a small amount of acetic acid, it is used again. The white lead 
is collected in (F), from which it is taken to be filtered and washed. 
The carbon dioxide used must be pure and concentrated, and is made 
by heating calcium carbonate with coke, in a special furnace (G). 
The gas is passed through water in (H) to remove impurities, and 
then goes to the precipitating vessel. 

The precipitation requires from 10 to 14 hours or more, varying, 
as does also the quality of the product, with the quantity and strength 

FIG. 87. 

of the solution of basic acetate. The white lead separates in a gran- 
ular or crystalline form, and is washed, ground, and dried, as in the 
methods already described. It is said to have less covering power 
than the amorphous powder produced by the Dutch method. 

Many chemical processes based upon the precipitation of a basic 
lead carbonate from solutions of various lead salts, have been pro- 
posed, such as Milner's process, the Kremnitz process, and cer- 
tain electrolytic processes, but these as yet have not developed 

Methods depending upon the precipitation of basic lead solutions 
with sodium carbonate have the disadvantage of forming crystalline 

Many electrolytic processes have been proposed : thus in modifi- 
cation of the chamber process, lead is placed on shelves, covered 
with carbon or tin plates, through which an electric current passes. 


More rapid corrosion of the lead thus charged, by the carbon dioxide 
and acetic acid vapors, with formation of a granular product, is 

It has been proposed to make a precipitated lead hydroxide by 
electrolysis with lead anodes and inert cathodes, in a bath of sodium 
nitrate, the hydrate being then treated with a solution of sodium bi- 
carbonate ; but this method of corrosion of the lead is expensive and 
has not proved a success. 

A general method for the electrolytic production of insoluble 
salts the formulae of which may be written as MA, in which M is the 
metal and A the anion of the salt, is to employ anodes of the metal 
M in an electrolyte containing two soluble salts, NA and NB, N being 
a suitable metal, usually an alkali, and B an anion whose salt with 
M is soluble. At the anode the salt MB is formed and diffuses out 
into the solution, where it is precipitated by double decomposition 
with NA. The precipitate not being formed on the anode itself is 
non-adherent and hence does not interfere with the electrolysis, as 
would otherwise be the case. A specific illustration of this method 
is the Liickow process.* The electrolyte, composed of solutions of 
80 parts sodium chlorate and 20 parts sodium carbonate, is diluted 
to such a degree that it contains 1.5 per cent of anhydrous salts, since 
dilute liquors yield the purest product. The liquid is kept slightly 
alkaline, and carbon dioxide is passed into it to replace the carbonate 
precipitated with the white lead. The cathodes of crude lead and 
the anodes of pure, soft lead, each having an area of 20 to 30 sq. dcm., 
are placed about 12 to 15 mm. apart, and the current density is 0.5 
ampere per sq. dcm. with a voltage about 2. The voltage varies 
somewhat according to the conductivity of the electrolytes and the 
distance between the pairs of electrodes. 

Owing to the high price of white lead, it is frequently adulterated 
with barytes (BaSCy, lead sulphate, lead carbonate, or calcium 
carbonate. Barytes is the most common adulterant, being cheap 
and heavy. A pure white lead should dissolve in dilute C.P. nitric 
acid, without leaving a residue. (Common nitric acid will not yield 
a perfect solution, as it contains sulphuric acid.) 

White lead is very heavy, having a specific gravity of 6.47. A 
cubic foot of the dry powder weighs about 185 pounds. It has 
great value as a pigment, owing to its covering power, its perma- 

* Mineral Industry, VIII, 392 ; IX, 438. J. Soc. Chem. Ind., 1895, 975 ; 1897, 


nency, and the readiness with which it mixes with other pigments. 
But it turns dark on contact with hydrogen sulphide, or if mixed 
with pigments containing sulphur, such as ultramarine, cadmium 
yellow (CdS), or vermilion (HgS). It is not suitable for painting 
the interiors of houses where gas or coal is burned. It is nearly 
insoluble in water, but, if taken into the system, will in time produce 
very dangerous poisoning; and too much care cannot be taken in 
the manufacture to prevent the fine dust from flying about. Sponges 
are worn over the mouth by the workmen, especially in the grinding 
room. An exhaust fan should be employed to draw the dust away 
from the workmen. 

Owing to the cost and poisonous character of white lead, substitutes 
are used to some extent. These are lead sulphate; sulphite, and oxy- 
chloride. Lead sulphate is the base of " sublimed white lead," the 
chief white lead substitute. By heating galena and coke in a blast of 
hot air, part of the lead is reduced to the metallic state, and part 
converted to sulphate and oxide, which, together with some metallic 
lead, sublime as " lead fume." This is collected in chambers and sub- 
jected to a second heating in a blast of hot air, which finishes the con- 
version to sulphate. The zinc present in the galena also passes off 
with the fume, and is converted to zinc white by the hot-air blast. 
The color of the sublimed white lead is sometimes improved by treat- 
ing with sulphuric acid. It has good covering power and color, and 
is not readily affected by hydrogen sulphide. It mixes well with other 
pigments containing sulphur, and is non-poisonous. 

Lead sulphite is made by precipitating a basic acetate solution 
with sulphur dioxide gas, or by subliming mixed lead and zinc ores 
with carbon, with a limited supply of hot air. 

Pattinson's white lead is an oxychloride of lead (PbCl OH), 
made by precipitating a hot solution of lead chloride with one-half 
the quantity of milk of lime necessary for its complete decomposi- 
tion. The pigment has good body and color, but is not now used. 

White zinc, or Chinese white, is zinc oxide (ZnO). It is made by 
distilling metallic zinc in fire-clay retorts, and leading the vapors 
into a flue through which air is drawn. On contact with the air, 
the hot vapor at once inflames and burns to the oxide, which is col- 
lected in a series of settling chambers, or in large bags of cotton 
cloth, the gas and air escaping through the meshes of the cloth. 
Instead of the metal, zinc ores mixed with carbon (e.g. coke or coal), 


are heated in special furnaces or retorts, the vapors being burned with 
air as before. But ores containing cadmium cannot be used, because 
cadmium oxide also sublimes, and being brown, discolors the product. 
The oxide is also formed by calcining zinc carbonate or hydroxide. 
The natural carbonate, Smithspnite, is, however, seldom pure enough, 
and precipitated carbonate must be used. This is -too expensive to 
compete with the combustion process. 

Zinc white is very permanent, and works well in water and in oil, 
of which latter it requires a large amount, usually about 20 per cent 
of its weight. 

Zinc sulphide is sometimes substituted for zinc white. This has 
more body than the oxide. If the vapors of zinc and sulphur are 
brought together, zinp sulphide is formed ; it is collected in settling 
chambers from which the air is excluded. As a rule, pure sulphide 
is not used, but a mixture of sulphide and barium or strontium 
sulphate. Zinc sulphide whites are permanent, have good body and 
color, and mix well with oil and with other pigments, excepting those 
containing lead or copper. 

Lithopone is a mixture of barium sulphate and zinc sulphide. 
Hot solutions of barium sulphide and zinc sulphate are mixed; the 
precipitate, after filtering and washing, is dried, ground with a little 
ammonium chloride, and the mass heated red hot and quenched by 
pouring into water. After grinding and levigating, a fine white 
powder is obtained, which works well in oil, has good body, is not 
readily affected by hydrogen sulphide, and is somewhat cheaper than 
white lead. It finds much use in paints and varnish enamels, for oil- 
cloths, and as filler in rubber compounding. 

The barium, sulphide required is made by reducing barytes with 
coal dust, by calcining in a rotary furnace or in a reverberatory. The 
charge is then lixiviated, hot, and the solution clarified by filtration. 

Barytes, or barium sulphate, occurs native in large quantities. 
The mineral is finely ground, treated with hydrochloric acid or with 
sulphuric acid to remove iron, and then levigated. Precipitated 
barium sulphate (blanc fixe) is obtained as a by-product in some 
chemical industries, and is used to a considerable extent as a filler 
and pigment. It has more body than barytes. 

Barytes is very heavy, is not affected by sulphur nor other chem- 
icals, and may be mixed with all pigments. It has little body, and 
does not work well in oil, having a streaky appearance when applied, 
and drying very slowly. Owing to its weight, one of its chief uses 
is to adulterate white lead. 


Gypsum, terra alba, or mineral white, is used to some extent 
as a pigment, especially for wall-paper printing. The mineral is 
simply ground, and treated with acid to remove the iron. Precip- 
itated calcium sulphate is a by-product of many chemical opera- 
tions, and is largely used as a filler in paper making, and for 
weighting cloth, under the names " Crown filler " and " Pearl 

Whiting, or Paris white, is calcium carbonate. It is prepared by 
grinding and levigating pure chalk, which occurs in large deposits 
in England, France, and other countries. Precipitated calcium car- 
bonate is a by-product from many chemical processes. Whiting is 
much used to modify the shade of other pigments, and as the bases 
of whitewash. When mixed with from 15 to 18 per cent of linseed 
oil, it forms putty. 

Kaolin, or China clay (p. 212), is sometimes used to modify the 
shade, or to adulterate other pigments. Its chief uses as pigment 
are in wall-paper printing, and as filler in cloth and paper. 


Ultramarine is the most important blue pigment. It occurs in 
nature as the mineral lapis lazuli, but in such small quantities, and 
the cost of preparation is so great, that this is of no importance as 
the source of the pigment. 

Ultramarine is probably a double silicate of sodium and alu- 
minum, together with a sulphide of sodium. But the composition 
varies in different samples having the same physical properties. 
The presence of sulphides seems necessary for the color, since, if 
treated with acid, hydrogen sulphide is evolved, and the color dis- 
appears. Numerous formulae have been proposed for ultramarine. 
Soda ultramarine, poor in silica, is 4(Na2Al2Si2Og) + Na2S4 ; * that 
high in silica is 2(Na 2 Al 2 Si3O 10 ) + Na 2 S 4 .* 

Soon after the introduction of the Leblanc soda process, blue 
spots, resembling natural ultramarine in color, were noticed in soda 
furnaces lined with siliceous material. This suggested the possi- 
bility of artificial ultramarine. In 1828, Guimet in France and 
Gmelin in Germany succeeded in making it. Guimet kept his method 
secret, but Gmelin published his. Afterwards, green, violet, and 
yellow ultramarine were discovered. A white ultramarine is sup- 
posed to be the basis of all others, and to it is assigned the formula : 

* Annalen der Chemie, 194, 1. R. Hoffmann. 


Na2Al 2 Si 2 O 8 + Na 2 S.* Green ultramarine is probably not a distinct 
chemical compound, but a mixture of ultramarines. 

None of the above ultramarines, excepting blue and green, has 
any commercial importance. 

The materials used for making ultramarines are China clay, 
sodium carbonate or sulphate, carbon, sulphur, and sometimes si- 
liceous matter. The purity of the material is important as affecting 
the shade of the color ; iron is especially liable to make it dull. There 
are two methods of making it, the sulphate of soda, or indirect method, 
and the soda-ash, or direct method. 

In the sulphate method, kaolin, anhydrous sodium sulphate, and 
charcoal, or pure coal, are powdered and thoroughly mixed. The car- 
bon is necessary to reduce the sulphate to sulphide. Sometimes rosin 
is used as a reducing substance. The kaolin should contain 2 SiO 2 to 
1 A1 2 O 3 , and be as finely powdered as possible. The mixture is packed 
in crucibles f having tight-fitting covers, and is heated at a bright red 
heat for about 8 hours. The furnace is allowed to cool very slowly, 
care being taken that no air has access to the contents of the crucibles. 
When cold, the mass is dull green and porous, and when ground and 
washed constitutes the ultramarine green of commerce. It is obtained 
by this process only. 

To make the blue ultramarine, the green powder is subjected to 
a " coloring " process. It is spread in shallow trays in layers about 
1 inch deep, and sprinkled with powdered sulphur. On heating, the 
sulphur ignites, and is allowed to burn itself out with access of air. 
Sometimes muffles are used, the sulphur being added in small quan- 
tities at a time, and the charge stirred with mechanical stirrers dur- 
ing heating. A part of the sodium sulphide is probably changed to 
the sulphate or other soluble salts, and the crude blue results. It 
is powdered and washed to remove soluble salts (Na 2 SO4, Na 2 SO 3 ), 
and sometimes boiled with a sodium sulphide solution to remove 
any free sulphur, which is injurious to the copper print rolls in 
calico printing. It is then ground and levigated, the different 
grades being used for different purposes. The shade is usually 
modified to match certain standards, by blending several lots of 

The soda-ash, or direct, method yields blue ultramarine at one 

* Annalen de Chemie, 194, 1. R. Hoffman. 

f In modern plants, muffle furnaces are replacing the crucibles for making the 
green ultramarine. But these must be built very carefully to exclude the air ; then 
hey need much time for cooling, usually requiring 10 days or more. 


heating, which may be done in muffles or in crucibles. The usual 
charge is about 2j tons, and consists of soda-ash, kaolin, charcoal, 
and sulphur, ground fine and packed firmly on the floor of the muf- 
fle, forming a layer about 14 inches thick. A layer of tiles, luted 
together with clay, is placed on top of the charge, and the front of 
the furnace is bricked up, a loose brick being left so that samples 
may be taken out to determine the time of heating. The process is 
very slow, requiring 3 or 4 weeks, of which 10 or 12 days are required 
for the slow cooling of the muffle; during all this time great care is 
necessary to exclude the air. The mass forms two layers, one bright 
blue, and the bottom greenish blue. These are separated, washed, 
and levigated. By using large crucibles instead of the muffle, the 
time of heating is reduced somewhat, but the breakage and extra 
labor more than offset the gain. 

To make an ultramarine which is less sensitive to acids, and which 
will withstand the alum used in paper making, a high percentage 
of silica in the pigment is necessary. For such a product, it is cus- 
tomary to use the soda process, and to add powdered quartz, sand, or 
diatomaceous earth to the charge. 

The first heating is very important in all processes of making 
ultramarine blue; about 700 C. is the proper temperature. If over- 
heated, the mass may fuse. Exclusion of air is necessary to prevent 
oxidation and loss of sulphur, which causes the product to turn dull 
green, brown, or gray. 

Ultramarine blue is much used in wall-paper and calico printing ; 
for neutralizing the yellow color in paper pulp, crystallized sugar, 
and cotton and linen goods ; for laundry blue ; for paint ; for printers' 
ink ; and for coloring mottled soaps. It is a very fast color to light, 
soap, and alkalies, but is quickly destroyed by even weak acids. 

Ultramarine violet is made by heating the blue, rich in silica, 
to 175 C., in an atmosphere of chlorine and steam. Some of the 
sodium is thus converted to salt, and removed by washing. The 
violet may also be formed by heating the blue to about 200 C., with 
2 or 3 per cent ammonium chloride, in the presence of air. It is not 
much used, as it has little tinctorial power. 

Ultramarine red is made by heating the blue to not over 145 C., 
in an atmosphere .of dry hydrochloric acid gas, or in the vapors of 
nitric acid. It is of but little importance. 

Prussian blue, or Berlin blue, is the ferrocyanide of iron (ferric- 
ferrocyanide), Fe 4 [Fe(CN) 6 ]3. To make it, a dilute solution of cop- 


peras (FeSO4 7 H^O), acidified with sulphuric acid, is precipitated 
with potassium ferrocyanide solution. After decanting the liquor 
the white precipitate of ferrous-ferrocyanide is oxidized with nitric 
acid, or with bleaching powder and hydrochloric acid. Exposure to 
the air also causes oxidation, but the color thus obtained is not so 

Chinese blue is a very pure and carefully prepared Prussian blue. 
In order to lighten the shade, and to make the pigment easier to 
grind, a certain amount of alum is added to the copperas solution 
before precipitating. 

A blue which is soluble in water results if the iron solution is 
poured into the ferrocyanide solution in a slow stream, or if Prus- 
sian blue is boiled in a ferrocyanide solution. In both cases, the 
ferrocyanide must be in excess. 

Prussian blue is not affected by acids, and mixes well with oil, 
but fades a little on exposure to the light. The color is destroyed 
by alkalies, and consequently it cannot be mixed with any sub- 
stance having an alkaline reaction. It has great tinctorial power, 
but is transparent, and lacks body. It is dissolved by oxalic acid, 
yielding a blue solution, formerly much used for blue ink. 

TurnbulTs blue, a deep reddish blue precipitate, is obtained by 
precipitating a ferrous salt with potassium ferricyanide [KsFe(CN)6], 
instead of the ferrocyanide. This is similar to Prussian blue. 

Smalt is a potash-cobalt glass, made by fusing pure sand and 
potash with cobalt oxide (Co 2 O 3 ), in a furnace similar to a glass fur- 
nace. The crude cobalt oxide, called " zaffre," is made by carefully 
roasting smaltite (CoAs2), cobaltite (CoAsS), or cobalt-nickel py- 
rites [(CoNi^Ss]. The ore is carefully sorted by hand, and iron 
pyrites and other impurities removed; then it is ground and some- 
times levigated, and roasted in a reverberatory furnace. A large 
part of the arsenic and sulphur passes off as oxides. The arsenic 
trioxide (As2O3) is condensed in long flues or chambers, while the 
sulphur dioxide escapes to the chimney. A small amount of the 
sulphur and arsenic is left in the zaffre to combine, during the 
fusion, with the iron, copper, nickel, and other injurious metals, form- 
ing a speiss, which, being heavier than the glass, settles to the bottom 
of the pot. The blue glass is refined (p. 203) until all the impurities 
have settled, and is then ladled out into water. This granulates 
it, and the sand so formed is ground under edge-runners and levi- 
gated. The medium-fine deposit is the best grade, the finest being 


too light-colored. The coarse and the very fine are usually re- 

Smalt is a very permanent color, fast to light, and not affected by 
acids nor alkalies. But it does not work well as a paint either in 
oil or in water, and is expensive; hence it is now largely replaced 
by ultramarine. The composition of commercial smalt varies much ; 
it may contain from 2 to 16 per cent of cobaltous oxide (CoO), but 
it is often difficult to get a good test for the cobalt. 

Imitation smalt is sometimes made of sand, colored with ultra- 
marine. A simple test with acid detects this at once. Prussian 
blue is shown by treating with alkali. 

Cobalt blue is made as follows : alumina is heated to a red heat 
in a crucible with basic cobalt phosphate, made by adding sodium 
phosphate to a cobalt nitrate solution. Alum and sodium car- 
bonate solutions are mixed, and aluminum hydroxide precipitated. 
These two products are thoroughly washed, and one part cobalt 
phosphate is mixed with 8 parts aluminum hydroxide, and the mix- 
ture heated to a red heat until the blue color develops. The pigment 
is then ground wet, washed, and dried. This yields a good oil color. 

Copper blues are not important. Mountain blue is the ground 
mineral azurite, a hydrated copper carbonate [2 CuCO 3 , Cu(OH) 2 ]. 

Bremen blue is a copper hydroxide containing some copper car- 
bonate and oxy chloride. A mixture of common salt, copper sul- 
phate, and metallic copper in small pieces is kept in tubs for several 
weeks, being well stirred frequently. A paste of green oxy chloride 
is formed, which is washed free from all soluble salts. A small 
quantity of hydrochloric acid is then added, and left for several 
hours. Finally, a solution of caustic soda is added, and thoroughly 
mixed until the paste acquires a blue color. After washing well 
and drying, it is ready for use. 

The copper blues are altered somewhat by exposure to the weather. 
They are readily darkened by hydrogen sulphide or sulphur fumes, 
so cannot be mixed with pigments containing sulphur. They dis- 
solve in acids and in ammonia, and become black when heated, 
owing to the formation of cupric oxide (CuO). They are opaque 
in water, but become slightly transparent in oil and lose body. They 
are at best a greenish blue. 

Indigo is an organic substance (p. 521) somewhat used as a pig- 
ment in laundry blue and soap. 



Ultramarine green is not largely used as a pigment. Its prepara- 
tion is described on p. 232. 

True Brunswick green is the oxychloride of copper, made by 
allowing metallic copper to stand for a number of weeks in a solu- 
tion of common salt which contains sulphates. The insoluble pig- 
ment is washed through a sieve to remove copper chips, and then 
dried at a low temperature to prevent decomposition. It is a good 
pigment, working well with oil, and having a fair coloring power; 
but the color is rather pale. 

The pigment now sold under the name of Brunswick green is 
generally a mixture of Prussian blue, chrome yellow, and barytes, 
the proportion of each depending on the shade desired. These 
greens are prepared by the dry or the wet methods. In the former, 
the dry ingredients are mixed in a paint- or edgerunner-mill. But 
the shade is inferior to that produced by the wet method. In this, 
copperas (FeSC>4 7 H^O), lead acetate, barytes, and potassium 
ferrocyanide and bichromate are used. The iron and lead salts are 
dissolved separately, and mixed while stirring in the barytes; some 
lead sulphate is thus precipitated also. Then, while still stirring 
actively, the mixture of potassium ferrocyanide and bichromate solu- 
tion is added. After a few moments' further stirring, the pigment is 
allowed to settle, and the liquor is decanted. Then the precipitate 
is washed by decantation, filtered, and dried carefully. Or the dry 
ingredients are finely powdered, and then stirred up thoroughly with 
water in a tank until, on settling, the liquor is nearly colorless. The 
precipitate is washed as above described. 

These greens are sometimes sold under the names Victoria, Prus- 
sian, or chrome green. They work very well in oil, have good cov- 
ering power, and are fairly permanent; but they cannot be mixed 
with pigments containing sulphur or alkaline substances, nor used 
where exposed to hydrogen sulphide gas. Alkalies act both upon 
the Prussian blue and the chrome yellow, causing them to turn red 
or brown. Sulphur darkens the chrome yellow. 

Chrome greens are valuable pigments, having a light yellowish 
green color. The basis is chromic oxide (Cr 2 O 3 ). By precipitat- 
ing a solution of a chromic salt with soda, chromium hydroxide 
[Cr(OH) 3 ] is obtained. This is washed, dried, and calcined at a 
red heat, until the water is expelled, and chromic oxide results. 


Guignet's green * is a chrome green made in the dry way. A 
mixture of 3 parts potassium bichromate with 8 parts boric acid 
is heated to dull redness in a reverberatory furnace for four hours. 
The porous mass is then washed, ground, and dried. In composi- 
tion, this green is a hydrated chromic oxide, containing a very small 
quantity of boric acid. A chromium borate is formed by the calci- 
nation, which is decomposed by the water, forming hydrated chro- 
mic oxide (Cr2O 3 2 H 2 O), or Cr 2 O(OH)4, and regenerating boric 

Guignet's green is permanent, mixes well with oil and with all 
other colors, and has good covering power. It is one of the most 
valuable pigments. 

Chrome greens, consisting of chromium phosphate, are sometimes 
made by boiling potassium bichromate with sodium phosphate and a 
reducing agent. These are not so good as the oxides, and have paler 

Copper greens containing only copper salts are of little impor- 
tance. Only two need be considered here. 

Mountain green, malachite, or mineral green, is a basic copper 
carbonate [CuCOs, Cu(OH) 2 ], occurring as the mineral malachite, 
which is much used for ornamental bric-a-brac and lapidary work. 
When ground very fine, it is sometimes used as a pigment, and is 
permanent in the light, misses well with oil, and has fair covering 
power. It is blackened by hydrogen sulphide. An inferior imita- 
tion of the natural product is made by precipitating copper sulphate 
solution with sodium or potassium carbonate containing a little white 
arsenic (As2Oa). 

Verdigris is not of constant composition, but is a basic copper 
acetate, corresponding nearly to trie-formula [2 Cu(C 2 H 3 O 2 )2 + CuO]. 
It is sometimes made by covering copper plates in heaps of the resi- 
due from wine presses. Fermentation of the mass produces acetic 
acid, which, together with the moisture, forms a layer of verdigris 
on the copper. This is scraped off, washed, and levigated. A better 
product is obtained by wetting cloths in vinegar or in pyroligneous 
acid, and spreading them between the copper plates. Verdigris is 
not a good pigment, being altered by moisture and light. 

By dissolving copper oxide, or carbonate, in acetic acid, and 
evaporating the solution, a crystallized salt having the composition 

* Bulletin de la Soci6t6 de Paris, 1, 9. Guignet, Fabrication des Couleurs, 


Cu(C 2 H 3 O 2 )2, Cu(OH) 2 H 2 O is obtained, which is called " distilled 
verdigris " in trade. This, however, is not a pigment. 

Copper and arsenic greens surpass all others in brilliancy and 
beauty, but, being exceedingly poisonous, cannot be used for many pur- 
poses. Scheele's green, which is chiefly copper arsenite (HCuAsO 3 ), 
is made by dissolving arsenious acid in a hot solution of potassium 
carbonate, and pouring the liquid into a solution of copper sulphate. 
The precipitate is carefully washed and dried. It is a grass-green 
pigment, having little coloring power, and now seldom used. 

Paris, or emerald, green is an aceto-arsenite of copper, 

[Cu(C 2 H 3 O 2 ) 2 CusAsfeOe], 

prepared by adding a thin paste of verdigris in water to a boiling 
solution of arsenious acid in water; some acetic acid is then added, 
and the mixture boiled until the precipitate is of the desired shade ; 
or the color will develop by simply allowing the mixture to stand 
for some days. By Galloway's process, sufficient sodium carbonate 
is added to a copper sulphate solution to precipitate one-fourth of 
the copper. Then acetic acid is added until the precipitate is just 
redissolved, and the liquid is heated to boiling. A hot solution of 
sodium arsenite (arsenious acid dissolved in sodium carbonate) is 
then added, and the mixture well stirred. The green precipitate is 
filtered, washed, and dried at a low ^temperature. For the finest 
pigment, the solutions should be dilute. 

Paris green has a peculiar light green shade possessed by no 
other pigment. It is permanent, works well in oil, and has a good 
covering power. But owing to its poisonous character its use as a 
pigment is much restricted. Nearly the whole of the present pro- 
duction is used to exterminate potato beetles and other insects inju- 
rious to vegetation. 

Terra verde is an earthy pigment, containing ferrous silicate as 
its chief ingredient. Green earths are found in numerous places, 
but the best are from Cyprus and Italy. They are a dull pale green, 
and are permanent, but have little covering power. 


The most important yellow pigments are chrome yellows and 
yellow ochres ; others are used but little. 

Chrome yellows have as a basis the chromate of lead, zinc, or 
barium, are all made by precipitation and each has a shade peculiar 


to itself. Lead chromate is made from the lead acetate, or nitrate, 
and potassium bichromate. The reactions are as follows : 

a) 2 Pb(C 2 H 3 02) 2 +K 2 Cr 2 07+H 2 = 2 KC 2 H 3 O 2 +2 C 2 H 4 O 2 +2 PbCrO 4 . 
6) 2 Pb(NO 3 ) 2 + K 2 Cr 2 O 7 + H 2 O = 2 KNO 3 + 2 HNO 3 + 2 PbCrO 4 . 

In order to modify the shade, lead, barium, or calcium sulphate is 
mixed with the chromate in the grinding-mill. Or a portion of the 
lead is precipitated as sulphate or carbonate along with the chro- 
mate ; this is done by mixing sodium carbonate or sulphate with the 
potassium bichromate. Chrome yellows are called " pure " when 
lead sulphate has been used to modify the shade. 

The precipitate is well washed by decantation, and the pulp 
freed from water in the filter press, or in a centrifugal machine, or 
by pressing in cloth bags. It should be dried at a low temperature, 
and well ground either dry or in oil. For the best color, the lead 
nitrate should be used in slight excess. When lead nitrate is used 
in making the chromate, it is customary to recover the potassium 
nitrate from the liquor and first wash-waters, the free nitric acid 
being neutralized with pearlash before evaporating. The excess of 
lead salt is precipitated from the waste liquors on the addition of 
the pearlash. 

Chrome yellow is sometimes made by digesting lead sulphate 
with a hot solution of potassium bichromate until the desired shade 
is developed. 

Lead chromate is a brilliant yellow, mixes well with oil, and has 
great covering power. It is blackened by hydrogen sulphide, and 
should not be mixed with pigments which contain sulphur, or are 
strongly alkaline. When treated with a caustic alkali, lead chro- 
mate is converted into a basic salt, having a red or orange color. 
These basic chromates are prepared for pigments, and sold under 
the name of chrome orange and chrome red. They are made by boil- 
ing chrome yellow with calcium or sodium hydroxide. The follow- 
ing is the reaction involved : 

2 PbCrO 4 + 2 NaOH =Na 2 CrO 4 + PbCrO 4 PbO - H 2 O. 

Quicklime gives a paler color than caustic soda. Chrome red is 
also made by digesting white lead with potassium bichromate and 
caustic soda. 

Zinc chromate is made from zinc sulphate and neutral potassium 
chromate. The neutral salt ZnCrO 4 forms only in concentrated 
solutions of the precipitant, and hydrolyzes instantly on contact 


with water to basic chromates, which precipitate, and to free chromic 
acid in solution. The composition of the precipitate varies and is 
determined by the concentration of the acid solution with which it is 
in contact. Indefinite washing with water will remove all of the 
chromic acid, but after the basicity has reached approximately 
4 ZnO CrOs, the loss of acid is slow, owing to the insolubility of the 
precipitate. Zinc chromate is also made by boiling zinc oxide with 
potassium bichromate. The pigment has a light lemon color, is 
permanent, not affected by sulphur, and can be mixed with other 
pigments. It is soluble in mineral acids, and is decomposed by caustic 

Barium chromate is much like the zinc salt, but is a greenish 
yellow color. It is made in the same way as is the zinc chromate, 
but from barium chloride. It is not used to any extent. 

Yellow ochres and Siennas are natural mineral products, varying 
from bright yellow to brown. The color is due to hydrated oxide of 
iron, and in Sienna there is a little manganese oxide. The pigments 
contain sand and clay in large quantities, and are decomposition 
products from iron-bearing minerals. The Siennas are usually finer 
grained and contain less gangue mineral than the ochres. They 
occur in beds in the earth, and the only preparation necessary is 
grinding and levigating. They are very permanent, mix well with 
oil and with other pigments, have good covering power, and are 
cheap. If ochres and Siennas are calcined, the water of hydration 
is removed from the ferric hydroxide, and the color becomes orange 
or red. Burnt Sienna, made by heating raw Sienna to a low red 
heat, is reddish orange in color. 

Cadmium yellow is cadmium sulphide (CdS), and is made by 
precipitating a cadmium solution with hydrogen sulphide. If the 
solution is strongly acid, the color becomes more nearly orange. 

It is a brilliant yellow, very permanent, and mixing well with 
oil and with other pigments, excepting lead and copper compounds. 
It is chiefly used as an artist's color. Sometimes cadmium yellow is 
made by using ammonium sulphides instead of hydrogen sulphide 
to precipitate the pigment ; but in this case free sulphur is present 
in the precipitate, and causes changes in the color when mixed for 

Orpiment is arsenic trisulphide (As 2 S 3 ). It is found native as a 
mineral, which is simply ground for pigment. It is also extensively 


made by precipitating a dilute solution of arsenious acid in hydro- 
chloric acid with hydrogen sulphide; or by subliming a mixture of 
arsenious acid and sulphur from a retort. The pigment obtained by 
either method is finely ground. 

Orpiment is a very bright yellow, mixes well with oil, and has 
good covering power ; but it is not permanent on exposure to light, 
and cannot be mixed with many other colors. It is also very poison- 
ous. It is sold under the name of royal yellow, or king's yellow. 

Litharge is lead monoxide (PbO), made by oxidizing metallic 
lead at a high temperature, in rotating cast-iron drums, heated by 
an external fire. The drums have shelves or ribs inside, which pick 
up the melted lead and cause it to fall in thin films through the cur- 
rent of air drawn in by a fan. It is not so important as a pigment as 
for the preparation of " boiled linseed oil " (p. 357). It is also exten- 
sively used in making lead glass and in pottery glazes. 

Another variety of lead monoxide, having a lighter yellow shade, 
is " massicot," which is formed by oxidizing lead at so low a temper- 
ature that no fusion of the product takes place. It is chiefly pre- 
pared for the manufacture of red lead (p. 242). 

Yellow lead oxide is also made by heating white lead. 

Gamboge is a gum-resin obtained from a tree (Garcinia Morella 
Desr.) of Siam. Incisions are made in the bark of the tree, and the 
sap is collected in bamboo receivers, in which the yellow resin is 
left on evaporation. Gamboge emulsifies with water, and is used as 
a water-color paint. It cannot be used as an oil paint except when 
mixed with alumina. 

Indian yellow, or purree, is made by heating the urine of cattle 
that have been fed with leaves of the mango tree, the color being 
produced by an excessive secretion of bile, which has passed into 
the urine. The pigment precipitates, and is pressed and dried; it 
consists of salts of euxanthic acid, an organic body. It is a bright 
yellow, but not permanent in the light, and is very expensive. 


Orange mineral is lead tetroxide (Pb 3 O 4 ), prepared by heating 
white lead in the presence of air. It is usually made from the scum 
which collects on the surface of wash-waters used in levigating white 

2 PbCO 3 , Pb(OH) 2 + O = Pb 3 4 + 2 CO 2 + H 2 O. 



In composition and properties it is similar to red lead (below), 
but has a slightly lower specific gravity (6.95). 

Chrome orange has been described in connection with chrome 
yellow (p. 239). 

Antimony orange is antimony trisulphide, made by precipitating 
a moderately concentrated solution of antimony chloride with hydro- 
gen sulphide. The precipitate is washed in dilute hydrochloric 
acid, and then levigated. It must be dried at a low temperature. 

It has a bright orange color in oil or water, is permanent and of 
good body, but is decomposed by alkalies. It is chiefly used for 
vulcanizing rubber, producing the red " antimony rubber " of com- 


Red pigments form a numerous and important group, containing 
some of the brightest and most permanent colors. 

Red lead is lead tetroxide (Pb 3 O 4 ), having the same chemical 
composition as orange mineral (p. 241), but differing in its physical 
properties. It is made by the direct oxidation of metallic lead. 
The process is carried on in two stages. In the first or " dressing " 
operation the lead is converted into massicot by heating with free 
access of air in a reverberatory furnace to a temperature just above 
that of melted lead (340 C.). The temperature must be very care- 
fully regulated, for if the massicot melts it passes into ordinary 
litharge, from which red lead cannot be made. As fast as a layer of 
oxide forms it is pushed to the back of the hearth with a " rabble " ; 
finally, the unoxidized lead is allowed to run off, and the massicot is 
raked out and cooled. It is pale yellow, of granular texture, and 
contains pellets of unoxidized lead. It is finally ground and levi- 
gated, and then transferred to the second or " coloring process " ; it 
is heated to a dull red heat in a muffle or reverberatory furnace with 
access of air. The mass is stirred frequently to assist the absorption 
of oxygen, and to develop the color. Samples are taken at inter- 
vals, until the desired shade is obtained, which usually takes from 
40 to 48 hours ; then the furnace is allowed to cool. The product is 
usually ground before packing for market. 

Red lead is somewhat variable in color, but is a good pigment of 
great covering power and brilliancy. It has a specific gravity of 
8.5. Chemically, it is regarded as a mixture of lead monoxide and 
peroxide (2 PbO + PbC^), but commercial samples vary some from 
this formula. When treated with dilute nitric acid, the monoxide 


dissolves, leaving the peroxide as a brown powder; this constitutes 
a test for red lead, since no other red pigment turns brown with nitric 

A large use of red lead is for glass making, for which a very pure 
grade is necessary. Owing to its oxidizing effect with linseed oil, it is 
extensively used, mixed with this oil, as a lute in plumbing and gas 
fitting. It is much used as a protective paint for iron and steel. 

Chrome red is a basic lead chromate (PbCrO 4 , PbO H 2 O), made 
by boiling chrome yellow with caustic soda or with lime, as described 
on p. 239. It is also made by boiling white lead with a solution of 
neutral potassium chromate. When the desired shade is developed, 
the pigment is washed, ground, and levigated. 

It is a fairly bright red, of good body, working well in oil. Like 
all lead pigments, it is darkened by sulphur and hydrogen sulphide. 
It is sold as Chinese red, American vermilion, and Victoria red. 

An imitation of chrome red is made by coloring white lead, orange 
lead, or barytes with some of the coal-tar dyes, especially with eosins. 

Red ochre is made by calcining ordinary ochre at a low red heat 
until more or less of the water of hydration is driven off. The shade 
depends on the time of heating, the longer the calcination the more 
purple the product. Red ochres are essentially ferric oxide with 
alumina, silica, and lime. The native oxides, hematite and limonite, 
are seldom used for pigment, being hard to grind. But in a few places 
soft deposits of hematite are found, which yield a pale red pigment 
without further treatment than grinding. These ochres are sold as 
Indian red, light red, Venetian red, etc. 

Iron reds are now being prepared in large quantities, chiefly as 
by-products from other manufactures. These are sold as rouge, 
colcothar, Venetian red, etc., and all contain ferric oxide as the col- 
oring matter. 

When fuming sulphuric acid is made by the dry distillation of 
copperas (p. 82), a residue of ferric oxide remains in the retort. This 
is ground, levigated, and sold as colcothar. It is nearly pure F^Oa. 

In the manufacture of galvanized iron or tinned ware, the rolled 
sheet iron is dipped into a bath of acid to dissolve any oxide from 
its surface before putting it into the bath of melted zinc or tin. These 
acid " dipping liquors " contain much iron, which is precipitated by 
adding soda-ash or lime, and used as pigment. If sulphuric acid is 
used in the dipping liquors, and is neutralized with lime, the precipitate 


consists of ferric hydroxide, with more or less calcium sulphate. By 
calcining, a light red pigment, called Venetian red, is formed. 

Many metallurgical operations yield liquors containing much 
iron, which is precipitated with lime, forming Venetian red. 

These iron reds are very permanent and valuable pigments. 
They work well in oil, mix with all other pigments, have very good 
body, and are cheap, but the color is not so bright as in some pigments. 
The covering power is, however, largely dependent upon the methods 
of precipitation and ignition, which are carefully guarded trade se- 
crets. The density of samples of practically identical composition 
may vary three- or four-fold. 

Vermilion is mercuric sulphide (HgS). It occurs in nature as 
the mineral cinnabar, but the pigment is now all made artificially. 
It is one of the brightest reds, and has been known for a long time. 
It is made in two ways, by the wet and by the dry process. In 
the wet process, 100 parts of mercury are ground with 38 parts of 
flowers of sulphur until thoroughly incorporated; then the mass is 
digested at about 45 C., with a solution of 25 parts caustic potash in 
150 parts water. The mixture is stirred frequently, and any water 
lost by evaporation is replaced. After 2 or 3 hours the mass becomes 
brown, and then gradually turns red. When the desired color is 
acquired, which usually takes about 8 hours, the pigment is at once 
washed by decantation, since further action of the potash dulls the 
color. The pigment is ground and dried carefully. The tempera- 
ture must be kept between 40 and 45 C., for if overheated it becomes 
brbwn. Solution of potassium or sodium polysulphide may be used 
instead of the potash. The brilliancy of the color may be increased 
by treating with hydrochloric or nitric acid. 

The dry methods yield the best product. The Dutch process con- 
sists in heating mercury and sulphur together in shallow iron pans 
until they combine to form a black mercuric sulphide (HgS, ethiops 
mineral). This is pulverized, and introduced into earthenware re- 
torts in small amounts at a time. The larger part of the black sul- 
phide sublimes into the upper part of the retort as a bright red powder. 
This is ground, washed, treated with acid, and levigated. 

Chinese vermilion is the finest quality, and its manufacture was 
long kept a secret. Now it is known to be made by a process simi- 
lar to the Dutch method, but owing to the patience and care exercised 
by the Chinamen a very fine product is obtained. 

Vermilion is a very heavy, opaque, and brilliant pigment. Owing 


to its weight, it settles out of the oil when used for paint, causing 
difficulty in applying it evenly. It is permanent, and not readily 
affected by acids and alkalies. When heated in a closed tube, it 
turns black, and finally sublimes unchanged, thus furnishing a good 
test for its purity. It is sometimes adulterated with red lead, iron 
reds, or carmine lakes, but these leave a brown or black residue when 
heated. Vermilion is very expensive. 

Vermilionettes are brilliant red pigments, produced by coloring 
neutral white bodies, such as barium sulphate, lead sulphate, or 
white lead with coal-tar dyes of the eosin class. The white base is 
stirred up with a solution of the dye, and lead acetate or alum is 
added, which precipitates the color upon the white base. Orange 
mineral is sometimes mixed with vermilionettes to brighten the 
color. These work well in oil, have good body, and are brilliant, 
but fade on exposure to the light. 

Realgar, the disulphide of arsenic (As 2 S 2 ), occurs in nature in 
small quantities as a brilliant red mineral which, when ground, fur- 
nishes a fine pigment. But the chief supply is obtained artificially 
by fusing together white arsenic (As2O 3 ) and sulphur in the proper 
proportions, or by distilling arsenical ores with sulphur. The crude 
product is remelted, and arsenic or sulphur added, as need be, to 
give the desired shade. As a pigment, realgar is subject to the same 
disad vantages as orpiment (p. 241). It is much used, however, in 
preparing " Bengal lights," and for unhairing hides for tanning. 

Antimony red, or antimony vermilion, is an oxysulphide of anti- 
mony, made by precipitating an antimony chloride solution with 
thiosulphate of soda. On heating the solution to 55 C., a red pre- 
cipitate separates. This is washed and dried at about 50 C. It is 
also prepared by dissolving tartar emetic in tartaric acid solution, 
mixing with sodium thiosulphate, and heating to 90 C. 

Antimony red is used for oil and water colors, and to some ex- 
tent in calico printing. It has good body, and is permanent if not 
mixed with alkalies or with alkaline vehicles. 

Carmine pigment belongs to the class of pigments, called " lakes," 
which are metallic salts of organic color acids. The coloring matter 
in carmine is the organic substance carminic acid (CnHisOio), ob- 
tained from the bodies of the cochineal insect. The lake is prepared 
by extracting the crushed insects with hot water, filtering, and adding 
a solution of alum or tin chloride, and cream of tartar. After stand- 


ing, the pigment precipitates. Or the lake may be precipitated at 
once by adding sodium carbonate to the mixed solutions. The extrac- 
tion is done in tinned copper vessels, and hard water is said to improve 
the color of the pigment. 

Carmine is a very bright scarlet, the tin salt being brighter than 
the aluminum. It works well in oil and as a water color, but fades 
on exposure to sunlight. It is soluble in strong caustic alkalies. 

Cochineal lake, crimson lake, Florentine lake, and others, are car- 
mine lakes containing a larger proportion of alumina or metallic 
base than does carmine. 

Madder lakes and Brazil-wood lakes are prepared by precipitat- 
ing extracts of these substances with alum and tin, by adding so- 
dium carbonate. They furnish red pigments of various shades, but 
lacking in covering power. 

Yellow lakes are made from fustic, Persian berries, or quercitron 
bark extracts, in the same way as the madder lakes are made. 

Many of the coal-tar dyes may be precipitated as lakes, and a 
great number of pigments are thus prepared. But many of them 
are deficient in covering power, and lack permanency on exposure 
to the light. 


Umbers are ochres containing more manganese than Sienna con- 
tains. They are complex mixtures of silica, alumina, iron, manganese, 
lime, and other matter. There are two varieties, raw and burnt. 
Raw umber has received no further treatment than grinding and 
levigating. Burnt umber has been calcined at a low, red heat, 
whereby more or less of the water of hydration of the iron oxide has 
been driven out, giving a darker shade to the product. The best 
umber comes from Cyprus, but many other localities furnish it in 
various shades. It is very permanent, has good covering power, and 
mixes well with all other pigments. It is not affected by acids nor 
alkalies, and is cheap. 

Vandyke browns are indefinite mixtures of iron oxides and or- 
ganic matter. They are obtained from certain bog-earth or peat 
deposits, or from ochres containing bituminous matter. They are 
also made artificially from charred organic substances, such as bark, 
cork cuttings, or bone dust. Mixtures of lampblack, yellow ochre, 
and iron oxide are also sold as Vandyke browns. These pigments 
are permanent, mix well with all other colors, and have good body. 


Sepia is an organic pigment obtained from the cuttle-fish (Sepia 
officinalis) , that secretes it as a dark liquid, to be discharged in the 
water to hide his movements when disturbed. It is contained in a 
small sac, which is removed and dried. To purify the pigment, it 
is dissolved in caustic soda, and the decanted solution is acidified with 
hydrochloric acid. The pigment thus precipitated is washed and dried. 

Sepia is a dark brown, fine-grained pigment, very permanent and 
capable of mixing with all other colors. It is chiefly used as a water 
color by artists. 


Black pigments nearly all contain carbon as the base. The most 
important is lampblack, which is the soot produced by the incom- 
plete combustion of organic substances, mostly of an oily or resinous 
nature. The knots and refuse from pitch pine and hemlock, the crude 
mineral oils, residues from petroleum refining, and the " dead oils " 
from coal-tar furnish much lampblack. The temperature of burning 
is low and the air supply limited, so that a large part of the carbon 
remains unconsumed and is deposited as soot in a series of chambers, 
through which the combustion products are led. Some oil may 
distil into the chambers, mixing with the product, increasing its 
liability to spontaneous combustion in storage, and lowering its value 
for paint, as it dries very slowly. 

Carbon black is made by burning natural gas * so that the flame 
impinges upon a rotating, horizontal iron plate. The sudden lower- 
ing of the temperature of the flame causes a deposit of carbon, which 
is removed from the plate as it rotates, by a fixed scraper. An au- 
tomatic conveyer carries the pigment to the grinding and sifting 
apparatus. The product is free from oily matter and mixes readily 
in water, but with difficulty in oil. It is much used for printing-ink, 
paint, rubber mixing, coloring cement mortar, etc. 

Carbon from different sources varies widely in both covering 
power and color. When obtained from hydrocarbons, the higher 
the molecular complexity of the substance burned, the less the density 
of the product, and the browner the shade, although many brown 
products grind black in oil. Ivory black is the densest and blackest 
pigment, though high-grade bone-black develops great brilliancy in a 

Ivory-black is made by heating the refuse from ivory working 
in closed retorts until all organic constituents are decomposed. The 

* Mining and Engineering World, 1911, Oct. 28. J. Soc. Chem. Ind., 1894, 128. 


retorts must not be opened until quite cold. The charred mass is 
ground fine, and yields the finest quality of black pigment with respect 
to deadness of the black when ground in oil. It is an intense black, 
but since it acts like bone-char on organic coloring matter, it cannot 
be mixed with most pigments of an organic nature. 

Bone-black is an inferior black, made from bones charred in a re- 
tort. When coarsely pulverized, it is extensively used for decol- 
orizing syrups and oils. It is finely powdered for pigment, and is 
much used in making leather blacking, where the calcium phosphate 
and carbonate in it are also of importance 

Charcoal from soft wood, ground very fine, is sometimes used as 
a pigment, and to mix with other blacks. It is not so soft and fine 
as lampblack. 

Graphite is employed as a pigment in pencils, crayons, and in 
stove-blacking. It also forms the basis of a protective paint for 
metal. It is a dull black, very inert and permanent. 

Manganese ores, such as pyrolusite (MnCy and hausmannite 
(Mn 3 O4), are sometimes powdered for pigments. But they act as 
" dryers " when used with oil, and are rarely used in paint. 

Black lake, made from logwood decoction and potassium bichro- 
mate with copper sulphate, is a blue black, but not permanent. 

Tannate of iron blacks, derived from tannin liquors, copperas, and 
alum, also fade on exposure to the light. 


Lehrbuch der Farbenfabrikation. I. G. Gentele, Braunschweig, 1880. 

Das Ultramarine. C. Fiirstenau, Wien, 1880. (Hartleben.) 

Die Erd- Mineral- und Lackfarben. Dr. Mierzinski, Weimar, 1881. 

Chemistry of Pigments. J.M.Thomson. Lecture before the Society for the 
Encouragement of Arts, Manufactures, and Commerce. London, 1885. 

Fabrication des Couleurs. Ch. Er. Guignet, Paris, 1888. 

Oel und Buchdruckfarben. Louis E. Andes, Leipzig, 1889. (Hartleben.) 

Die Fabrikation des Ruses und der Schwaerze. H. Koehler, Braun- 
schweig, 1889. 

The Chemistry of Paints and Painting. A. H. Church, London, 1890. 

Painters' Colours, Oils, and Varnishes. G. H. Hurst, London, 1892. 

Pigments, Paints, and Painting. George Terry, London, 1893. (Spon.) 

Die Fabrikation der Mineral- und Lackfarben. J. Bersch, Leipzig, 1893. 

Die Fabrikation der Erdfarben. Dr. Josef Bersch, Leipzig, 1893. 

Handbuch der Farben-Fabrikation. Dr. S. Mierzinski, Leipzig, 1898. 

Chemistry and Technology of Mixed Paints. M. Tpch, New York. 

Modern Pigments and their Vehicles. Frederick Maire, New York, 1908. 

An Introduction to the Chemistry of Paints. J. N. Friend, London, 1910. 

Paint Technology and Tests. H. A. Gardner, New York, 1912. 

Journal of American Chemical Society, 1880, 381. H. Endemann. 

Journal of the Society of Chemical Industry : 

1887,719. Rawlins. 1890,1137. Wunder. 1891,709. 1892, 357. Weber. 

Jour. Ind. Eng. Chem., 1914, 54. 



Bromine is widely distributed in nature as bromides, usually 
accompanying common salt and magnesium chloride. The world's 
supply is obtained from " bittern," the mother-liquor of the salt 
industry. Stassfurt furnishes about two-thirds of the supply, and 
the remainder is extracted from the brines found in Michigan, Ohio, 
and West Virginia, along the Kanawha and Ohio rivers. The 
American product in 1913 was about 572,400 pounds. Small quan- 
tities are obtained from the mother-liquors of the Chili saltpetre 
industry, and in Europe from kelp. 

Bromine is present in the mother-liquors as magnesium bromide, 
and to a small extent as sodium bromide; the liquors also contain 
large quantities of sodium and magnesium chlorides. Several methods 
of extraction are in use, the continuous and periodic processes 
being old, while recently direct electrolysis of the waste brine has 
been introduced. The bromine is liberated by the current before the 
chlorine is set free. 

The continuous process depends on the decomposition of the 
magnesium bromide by chlorine gas. A sandstone or earthenware 
tower is filled with broken brick or burned clay balls; chlorine gas 
and steam are introduced at the bottom of the tower, and rising 
between the balls, meet descending streams of hot bittern. By reac- 
tion between the chlorine and the magnesium bromide, the bromine 
is set free. The chlorine stream must be regular, and so controlled 
that no excess is used; otherwise some bromine chloride is formed. 
Part of the bromine dissolves in the liquor as soon as set free, and 
this liquor flows into a special receiver, heated by steam ; here it is 
boiled to drive out the bromine, which, together with water vapor, 
passes back into the tower, entering at the bottom, and mixing with 
the chlorine. At the top of the tower, the bromine vapor passes out 
into an earthenware worm-condenser, which empties into a closed 
vessel. An outlet pipe from the top of this receiver passes into 
a small tower, filled with moist iron turnings or scrap iron. Any 
uncondensed vapors of bromine, passing out of the receiver, combine 
with the iron to form ferrous bromide, which is used for making 
potassium bromide. 

In this process, any bromine chloride formed in the tower is decom- 
posed, before it can pass into the condensing worm, by the fresh 



bittern entering at the top of the tower. Bromine chloride is a vol- 
atile liquid, and would contaminate the bromine. The exhausted 
bittern from the heating-vessel goes to waste. The chlorine gas 
necessary is made in special stills from manganese binoxide and hydro- 
chloric acid. 

The periodic process depends on the following reaction : 

MgBr 2 + 2 H 2 SO 4 + MnO 2 = MgSO 4 + MnSO 4 + 2 H 2 O + 2 Br. 

This is carried on in sandstone stills, heated by steam. A charge 
of pyrolusite, sufficient for several days, is put into the still, and 
the bittern, heated to 60 C., is run in. The quantity of sulphuric 
acid to be added is carefully gauged with each charge of bittern in 
order that none of the magnesium chloride shall be decomposed. 
Usually, a little of the magnesium bromide is left in the bittern, 
since the high temperature necessary to decompose the last traces 
would also decompose some of the chloride, which would form bro- 
mine chloride, and contaminate the product. The bromine distils 
over into a condensing worm, as above described. The exhausted 
bittern is drawn off after each charge, and goes to waste. At the 
present time, considerable potassium chlorate is used instead of 
pyrolusite for the oxidizing agent. This is especially advantageous 
if the bittern contains much calcium chloride, since only one-half as 
much sulphuric acid is necessary, and there is, consequently, less 
difficulty from calcium sulphate. Neither the stills nor the tower 
should be lined with pitch or tar, since these substances absorb much 

The crude bromine obtained by either process contains some bro- 
mine chloride, lead bromide from the pipe-joints and connections, 
and some organic matter. It is purified by shaking with ferrous, 
sodium, or potassium bromide, and redistilling from glass retorts. 
The bromine chloride is thus decomposed, and the salts of the heavy 
metals remain in the still. Very pure bromine is obtained by neu- 
tralizing with barium hydroxide solution, evaporating to dryness, and 
calcining at a red heat. The barium bromate and chlorate formed in 
the neutralizing are decomposed to form barium bromide and chlo- 
ride. By extracting the mass with alcohol, the bromide is dissolved. 
The barium bromide obtained by evaporation of the alcohol is de- 
composed with pyrolusite and sulphuric acid, the pure bromine 
passing to the condenser as vapor. 

Operations with liquid bromine must be carried on in the open 
air, or in a strong draught. If inhaled, the vapors are suffocating, 


and cause great irritation of the air passages. The liquid attacks 
the skin, and causes sores which heal very slowly. 

Bromine is largely used in making certain coal-tar dyes, such 
as the eosins ; for sodium and potassium bromides ; and to some 
extent as a chemical reagent, and for making organic bromides. It 
is considered dangerous freight by transportation companies, and so 
only its salts, especially potassium bromide, are usually shipped. 

" Solidified bromine " is a convenient form for laboratory work. 
This consists of sticks of diatomaceous earth, pressed with size or 
molasses, burned till coherent, and soaked in liquid bromine. The 
porous material absorbs from 50 to 75 per cent of its weight of the 

Potassium bromide is made by decomposing iron bromide with 
potassium carbonate. The ferroso-ferric bromide (Fe 3 Br 8 ), made by 
adding more bromine to ferrous bromide, is usually employed. 

Fe + Br 2 = FeBr 2 . 

3 FeBr 2 + 2 Br = Fe 3 Br 8 . 

Fe 3 Br 8 + 4 K 2 CO 3 + 4 H 2 O = Fe 3 (OH) 8 + 8 KBr + 4 CO 2 . 

The solution is filtered and evaporated, yielding cubical crystals of 
the salt, free from bromate, which is always formed when bromine 
is neutralized directly with alkali. 

Potassium bromide is used in medicine and in photography, espe- 
cially in the preparation of silver bromide plates and films. 

Sodium bromide is similar to the potassium salt, is used for the 
same purposes, and is made in the same way ; but it does not crystal- 
lize so well. 


Berichte iiber die Entwickelung der chemischen Industrie. A. W. Hof- 

mann, 1875, 129. 

Moniteur scientifique, 1879, 905. H. S. Welcome. 

Handbuch der Kali-Industrie, E. Pfeiffer, 321. Braunschweig, 1887. 
Die Gewinmmg des Broms in der Kaliindustrie. M. Mitreiter, Halle, 

a. S., 1910. 

Salt Deposits in Ohio. Bull. 8, vol. IX (1906), Rep't. Ohio Geol. Survey. 
J. A. Bownocker. 


Iodine is obtained from the ashes of seaweed, and from the mother- 
liquors of the Chili saltpetre industry. 

Along the coasts of France, Scotland, and Norway, seaweed is 
collected and burned * at as low a temperature as possible. The ash, 
called kelp, or varec, contains from 0.5 to 1.5 per cent of its weight 
of iodides of sodium and potassium. It is lixiviated, and the filtered 
solution is systematically evaporated. First, sodium sulphate, and 
then common salt, crystallizes. By further evaporation, sodium 
carbonate, together with more salt and potassium chloride, sepa- 
rates. The mother-liquor is then treated with sulphuric acid to 
decompose the alkali sulphides and sulphites formed by reduction of 
the sulphates during incineration. This precipitates sulphur, and 
the sodium sulphate formed crystallizes. The mother-liquor, still 
holding the iodides in solution, is then heated to 60 C. in iron re- 
torts with lead covers, and having pipes leading to condensers.! 
Small quantities of pyrolusite are introduced into the retort period- 
ically, when the following reaction takes place : 

2 Nal + 3 H 2 S0 4 + MnO 2 = MnSO 4 + 2 NaHSO 4 + 2 H 2 O + I 2 . 

Pyrolusite is added as long as iodine distils off ; but excess must 
be avoided, lest bromine and chlorine be set free from the salts still 
present in the liquor, and combine with the iodine to form tribrom- 
or trichlor-iodine (IC1 3 ). 

Sometimes the iodine liquor is decomposed by leading chlorine 
gas into it, the same as in making bromine (p. 249). The crude 
iodine precipitates as a paste, and is washed and then dried on porous 
plates. Much care is necessary to avoid an excess of chlorine, since 
this forms volatile iodine trichloride (IC1 3 ), and causes loss. 

By heating the acidified iodide solution with ferric chloride or 

* By burning the seaweed in closed retorts, the loss of iodine by volatilization is 

t The condensers, called udells, are bottle-shaped vessels of earthenware, ar- 
ranged horizontally, 5 or 6 in a series, the neck of one entering the bottom of the 
next. In the lower side of each is a small hole, through which the condensed water 
drains off. Each still has two sets of udells, which are left in position during re- 
peated charges of the still, until they are filled with solidified iodine. Recently 
the condensers have been made of seven or eight lengths of plain earthenware pipe, 
each length 3 feet long by 1| feet in diameter, and the joints luted with clay. 



potassium chlorate, the iodine is liberated, and distils off, with 
some water, and no trichloride is formed, thus : 

a) 2 Nal + 2 FeCl 3 = 2 FeCl 2 + 2 NaCl + I 2 . 

6) 6 Nal + KC10 3 + 3 H 2 O = 6 NaOH + KC1 + 6 I. 

Another method is to mix the kelp with a little water and sul- 
phuric acid, and to add potassium bichromate : 

6 Nal + 10 H 2 SO 4 + K 2 Cr 2 O 7 = 6 NaHSO 4 + K 2 Cr 2 (SO 4 ) 4 + 7 H 2 O+ 6 1. 

The precipitated iodine is washed, dried, and sublimed. 

It has been proposed to heat the kelp directly with powdered 
bichromate, decomposition taking place at a red heat, and the iodine 
subliming : 

6 KI + K 2 Cr 2 O 7 = 4 K 2 O + Cr 2 O 3 + 61. 

The seaweeds of the Pacific coast of America may also furnish 
iodine in large quantity, when the market conditions will warrant its 

The recovery of iodine from the mother-liquors of Chili saltpetre 
is now most important. The iodine is chiefly in the form of sodium 
iodate (NalOs), and the process depends on the following reaction : 

2 NaIO 3 + 5 SO 2 + 4 H 2 O = Na 2 SO 4 + 4 H 2 SO 4 + I 2 . 

In practice, the sulphur dioxide is used in the form of sodium 
bisulphite solution, containing some neutral sulphite. This is made 
immediately before use by leading sulphur dioxide gas into sodium 
carbonate solution until the liquid contains one part of neutral sul- 
phite to two of acid sulphite. The requisite quantity of this acid 
sulphite liquor is added to the mother-liquor, and thoroughly agi- 
tated; the precipitated iodine is collected on filters made of coarse 
bagging or canvas, and after washing is pressed heavily to remove 
excess of water. The reaction is probably as follows : 

2 NaIO 3 + 3 Na 2 SO 3 + 2 NaHSO 3 = 5 Na 2 SO 4 + I 2 + H 2 O. 

But since some sodium iodide is also present, the excess of bisul- 
phate employed decomposes it according to the reaction : 

NaI0 3 + Nal + 2 NaHSO 3 = 2 Na2SO 4 + I 2 + H 2 O. 

Sometimes the iodine is precipitated as cuprous iodide (Cu 2 I 2 ) by 
adding copper sulphate and sodium bisulphite to the mother-liquors, 


but this is now less frequently done than formerly. The cuprous 
iodide was shipped to Europe, and used to make potassium iodide 
by treating with potassium carbonate. 

The liquors from which the iodine has been separated are re- 
turned to the lixiviation tanks for the treatment of the crude " caliche " 
(p. 145). 

The crude iodine obtained by any of the above processes is puri- 
fied by resubliming in iron retorts, the vapors being condensed in 
earthenware receivers. The temperature of the retorts must be very 
low in order to form large crystals, and the condensers must not be 
so cool as to cause sudden condensation of the vapors. 

The chief uses of iodine are in the manufacture of coal-tar dyes 
and organic compounds, and in medicinal preparations. 

The most important iodine derivative is potassium iodide (KI). 
This is made in several ways : 

(a) Iodine may be dissolved in a caustic potash or carbonate 
solution, the solution evaporated to dryness, and the mixture of 
iodide and iodate so obtained calcined with powdered charcoal at a 
low red heat, to decompose the latter salt. 

61 + 6 KOH = 5 KI + KIO 3 + 3 H 2 O. 
2 KI0 3 + 3 C = 2 KI + 3 

The calcined mass is lixiviated, filtered, and crystallized. Very pure 
materials are needed in this process. 

(b) A better method is to form ferroso-ferric iodide, and decom- 
pose this with pure potassium carbonate. Metallic iron is dis- 
solved by digesting with iodine and water, forming ferrous iodide, 
which is then treated with sufficient iodine to form the ferroso- 
ferric salt : 

Fe + 2 I = FeI 2 . 

3 Fel 2 + 21 = Fe 3 I 8 . 

Fe 3 I 8 + 4 K 2 CO 3 + 4 H 2 O = Fe 3 (OH) 8 + 8 KI + 4 CO 2 . 

This method, if carefully worked, yields a very pure salt, entirely 
free from potassium iodate. The precipitated ferroso-ferric hydrox- 
ide is granular, and more easily washed than is ferrous hydroxide. 

(c) Barium iodide is made by agitating barium sulphide solution 
with iodine. The clear solution is then boiled with potassium sul- 
phate solution, the precipitated barium sulphate filtered off, and the 
filtrate evaporated until the potassium iodide crystallizes : 


I 2 = BaI 2 +S. 
BaI 2 + K 2 SO 4 = BaSO 4 + 2 KI. 

Potassium iodide is chiefly used in medicine as an alterative and 
diuretic. A small quantity is used in photography. 

Lead, mercury, and ferrous iodides are used to a small extent in 
medicine, but these are not important. 


Wagner's Jahresbericht uber die Leistungen der chemischen Technolo- 

S'e, 1879, 334. E. Sobering. (Jodkalium.) 337. G. Langbein. (Jod- 
ewinnung in Chili.) 
Journal of the Society of Chemical Industry : 

1893, 128. J. Buchanan. (Extraction of Iodine in Chili.) 


The discovery of phosphorus, about 1675, is attributed to an 
alchemist, Brand, at Hamburg. Urine which had been evaporated 
to a thick syrup was heated in an earthenware retort with sand, the 
phosphorus distilling off. It was known only as a chemical curios- 
ity until Scheele, in 1775, made it from bone-ash ; soon after it as- 
sumed commercial importance. Bone-ash is still a leading source, 
but the mineral phosphates, being cheaper, are now used. 

Normal calcium phosphate [Ca 3 (PO 4 )2] is reduced by carbon 
only at excessively high temperatures, and then forms calcium phos- 
phide rather than the free phosphorus. Free phosphoric acid is, how- 
ever, reduced to phosphorus. In the old process, tricalcium phos- 
phate (as bone-ash) was decomposed with sulphuric acid to form 
monocalcium phosphate and calcium sulphate. By leaching with hot 
water, the monocalcium phosphate was dissolved, and the solution, 
after decantation from the sulphate, was evaporated in lead pans, 
when powdered charcoal or coke was stirred in, and the mass heated 
in iron pans until dry. The dry mixture was charged into earthen- 
ware retorts and heated moderately at first, and then to very high 
temperatures. The moderate heating reduced the monocalcium phos- 
phate to calcium metaphosphate, which in turn was decomposed by 
the carbon, forming tricalcium phosphate, phosphorus, and carbon 
monoxide. The reactions were as follows : 

Ca 3 (PO 4 ) 2 + 2 H 2 SO 4 = CaH 4 (PO 4 ) 2 + 2 CaSO 4 . 

CaH 4 (P0 4 ) 2 = 2 H 2 O + Ca(PO 3 ) 2 . 

3 Ca(PO 3 ) 2 + 10 C = Ca 3 (PO 4 ) 2 + P 4 + 10 CO. 

This left one-third of the phosphorus combined with the calcium, 
but by adding silica to the mixture, all of the phosphorus is liberated, 
according to the reactions : 

Ca(PO 3 ) 2 + 5 C + SiO 2 = CaSiO 3 + 2 P + 5 CO. 
Ca 3 (PO 4 ) 2 + 5 C + 3 SiO 2 = 3 CaSiO 3 + 2 P + 5 CO. 

The electric furnace process of Headman,* Parker, and Robinson, 
for the direct reduction of calcium phosphates in a continuous-acting 

* J. Soc. Chem. Ind., 1891, 445. U. S. Pat. No. 482,586 (1892). 



FIG. 88. 

furnace (Fig. 88) ha? now replaced the older methods. The retorts 
employed in the old process, to withstand the chemical action of the 
charge, had to be made from materials 
which are poor conductors of heat, and hence 
the wear and tear was heavy, and the heat 
efficiency low. Electrical heating develops 
the energy within the retort itself, and 
allows the retort walls to be kept rela- 
tively cool. Owing to the high tempera- 
ture attained by this method of heating, 
silica (as sand) may be introduced directly 
in the charge, all of the phosphorus being 
set free to distil out of the furnace, while 
a fused slag is separated and tapped off at the base of the furnace. 

An intimate mixture of carbon, phosphate, and flux is heated ; the 
gases and phosphorus vapors pass by the pipe (P), to the condenser, 
while slag is tapped off at intervals, through (0). Fresh charges 
are introduced through (H), by the conveyer (C). The carbon 
electrodes (E) are in metal sockets passing through the walls of the 
furnace. The working holes (X) are closed with clay when the fur- 
nace is running. This method avoids the use of sulphuric acid, the 
concentration and handling of phosphoric acid, uses no earthenware 
retorts, and saves time ; it is further claimed that less coke is used. 

The crude phosphorus made by any of the above processes con- 
tains sand, carbon, clay, and other impurities. It is purified by 
melting under warm water, and straining through canvas bags ; 
formerly chamois leather was used. Or it is redistilled from iron 
retorts. Sometimes it is treated with a 3 per cent solution of potas- 
sium bichromate and its equivalent of sulphuric acid, in a lead-lined 
agitator which is heated by steam coils. After a couple of hours' 
agitation, the phosphorus is nearly transparent, and of a light yel- 
low color. It is washed with hot water, filtered through canvas 
bags, and moulded into " sticks " by pouring into glass or tin tubes 
placed in cold water. For shipment, phosphorus is packed in water 
in tin boxes, the lids of which are tightly soldered. 

Yellow or ordinary phosphorus is a pale yellow, translucent, wax- 
like mass of 1.82 specific gravity, very inflammable, and combining 
directly with oxygen, sulphur, and the halogens. It melts at 43.3 
C. under water, and at 30 C. when dry; it distils at 269.* It is 
very soluble in carbon disulphide, sulphur chloride, and phosphorus 

* J. B. Readman, Thorpe's Dictionary of Applied Chemistry, Vol. IV., 205. 



trichloride, slightly so in caustic soda solution, but insoluble in 
water. It is exceedingly poisonous, less than 0.15 gram being a 
fatal dose. Persons working continuously with yellow phosphorus 
are subject to necrosis, usually appearing first in the jawbones. 

The chief uses of yellow phosphorus are in making matches and 
phosphor-bronze, and for rat poison. 

Amorphous or red phosphorus is made by heating the yellow 
variety for several hours in closed retorts at 250 C. If an auto- 
clave be employed, and the temperature raised to 300 C., the press- 
ure inside the vessel makes the process much more rapid. The 
hard mass thus produced is ground under water, and the powder 
boiled with caustic soda solution to remove any unchanged yellow 
phosphorus. Carbon disulphide is sometimes used instead of caus- 
tic soda, but this is expensive and easily inflamed. After boiling in 
water, filtering, and drying by steam heat, the amorphous phospho- 
rus is packed dry." Red phosphorus is a reddish brown, opaque 
substance, having a specific gravity of 2.25. It is not affected by 
heating in the air until the temperature reaches 260 C., at which 
point it inflames. By heating in an atmosphere of nitrogen or car- 
bon dioxide, it distils, returning to the yellow variety. It is insol- 
uble in carbon disulphide, caustic soda, and in water, and is not 
poisonous. The chief use of red phosphorus is in the manufacture 
of " safety matches." 


In about 1812, the so-called "chemical matches " were invented. 
Sticks were dipped in melted sulphur, and the "head " coated with 
a mixture of sugar and potassium chlorate. It was fired by dipping 
into a bottle containing asbestos moistened with sulphuric acid. 

Friction, or lucifer, matches were invented in 1827, in England. 
The heads were a mixture of antimony trisulphide and potassium 
chlorate, made into a stiff paste with water and gum. They were 
ignited by rubbing on sand or emery paper. The antimony trisul- 
phide was soon replaced by phosphorus, and the potassium chlorate 
by nitre. At the present time, lead peroxide, red lead, or man- 
ganese dioxide are used instead of nitre as the oxidizing substance. 
Chlorates are used; but sparingly, since they form explosive mixtures. 

Soft wood, generally pine or spruce, is cut by machines to form 
the sticks, which are thoroughly kiln-dried. They are then fixed in 
a frame so that each stick stands alone, and the end of each stick is 
well soaked in melted sulphur, paraffine, or stearic acid. The igniting 



mixture is made by ,slowly stirring phosphorus into a warm solution 
of dextrine or glue ; the oxidizing materials are then added, and the 
paste stirred until cold. It is frequently colored with ultramarine, 
lead chromate, chalk, or lampblack. It is then spread evenly in a 
thin layer on a table, and the prepared sticks dipped into it once or 
twice. After drying, the heads are sometimes dipped in thin shel- 
lac or other varnish, to protect them from the moisture in the air. 

Safety matches are made without yellow phosphorus. The match 
head is generally sulphur, or antimony trisulphide, with potassium chlo- 
rate, or bichromate, as the oxidizing material. Sometimes red lead, lead 
peroxide, or manganese dioxide is used as a part of the oxidizing mate- 
rial. The surface upon which the match must be lighted is coated with 
a mixture of red phosphorus, antimony trisulphide, and dextrine, or 
glue. Powdered glass or emery is used to increase the friction. 

The compositions used on matches are carefully guarded as trade 
secrets, and are different in different factories. One is given as 
follows : 


KC1O 3 5 parts 

K 2 Cr 2 O 7 2 parts 

Glass Powder .... 3 parts 

Gum 2 parts 


Sb 2 S 3 5 parts 

Red Phosphorus ... 3 parts 

MnO 2 1 parts 

Glue 4 parts 

The friction of the match head on the prepared surface develops 
sufficient heat to convert a little of the red phosphorus to the yellow 
variety, which at once combines with some of the potassium chlorate 
and antimony sulphide, evolving enough heat to inflame the mix- 
ture on the head. 

To prevent the burned stems from smouldering, the sticks are 
sometimes soaked in a solution of magnesium sulphate, alum, or 
sodium phosphate before making the head. 

In some countries, e.g. Switzerland, the manufacture and sale of 
matches containing yellow phosphorus is prohibited. . In this country 
their production has been eliminated by the imposition of a prohib- 
itive tax of 1 cent per 100 matches. 


Chemical News, 1879, 147. J. B. Readman. (Manufacture of Phos- 

Chemiker-Zeitung, 1881, 196. A. Rossel. (Matches without Phosphorus.) 
Journal of the Society of Chemical Industry : 

1890, 163, 473. J. B. Readman. (Manufacture of Phosphorus.) 

1891, 445. J. B. Readman. (Manufacture of Phosphorus.) 


Boric acid [B(OH) 3 ] occurs in volcanic regions, especially in Tus- 
cany, as a constituent of the vapors, called soffioni, which escape 
from hot springs and from openings in the ground, called fumeroles. 
In some places the water has evaporated from the fumeroles, and the 
boric acid has crystallized, forming the mineral sassolite. Combina- 
tions of boric acid with sodium, magnesium, and calcium are found 
in various places : as, tinkal (native borax), Na 2 B 4 O 7 10 H 2 O ; bora- 
cite, 2(Mg 3 B 8 Oi 5 ), MgCl 2 ; borocalcite, CaB 4 O 7 6 H 2 O ; and borona- 
trocalcite (ulexite), Na 2 B 4 O 7 , (2 CaB 4 O 7 ), 18 H 2 O. 

In Tuscany, natural or artificial ponds (lagoons) are formed 
around the fumeroles, or a series of masonry basins or tanks are 
constructed over them, and the soffioni made to bubble through 
water in these, thus washing most of the boric acid from the vapors. 
These tanks are so arranged that the water from one flows into an- 
other at a lower level; in the final basin, a solution containing about 
2 per cent boric acid is obtained. The solution is evaporated, either 
in lead-lined vessels, heated by the steam from the fumeroles, or in 
cement-lined tanks, having coils through which the steam passes. 
Calcium sulphate deposits freely during the evaporation of the 
solution, which is concentrated to 1.08 specific gravity. It is then 
crystallized in lead-lined wooden vats. The crystals are drained 
for some hours, and dried on a floor also heated by steam from the* 
fumeroles. The crude boric acid is purified by recrystallization. In 
many places in Tuscany, bored wells are sunk from 200 to 300 feet, 
and the vapors escape from these as from the natural fumeroles. 

Boric acid is made in California, and in Chili, by boiling calcium 
borates, suspended in water, with sulphur or sulphurous acid : 

Ca 2 B 6 On 5 H 2 O + 8 S + 10 H 2 O = 2 CaSO 4 + 6 B(OH) 3 + 6 H 2 S. 
CasBAi 5 H 2 + 4 SO 2 + 6 H 2 O = 6 B(OH) 3 + 2 Ca(HSO 3 ) 2 . 

Much boric acid is made from the boracite in the Stassfurt salts. 
The mineral is crushed, and treated with just enough hydrochloric 
acid to decompose it. A rather vigorous reaction takes place, and 
the mass becomes pasty. It is dissolved in boiling water, and care- 
fully tested for free hydrochloric acid; if none is present, the solu- 
tion of boric acid is decanted from the sediment of clay and sand, or 



filtered through linen bags, and is crystallized in lead-lined or iron 
tanks.* Sulphuric acid is also used to decompose the boracite, in 
which case the mother-liquors from the boric acid contain magne- 
sium sulphate; this is recovered as Epsom salt. The following 
reactions are involved : 

1) (2 Mg 3 B 8 Oi 5 ), MgCl 2 + 12 HC1 +. 18 H 2 O = 7 MgCl 2 + 16 B(OH) 3 . 

2) (2 Mg 3 B 8 15 ), MgCl 2 + 7 H 2 SO 4 + 18 H 2 O = 

7 MgSO 4 + 2 HC1 + 16 B(OH) 3 . 

The actual quantity of acid used is determined for each lot of salt. 

Boric acid forms pearly white, laminated crystals, very slightly 
soluble in cold water, but dissolving readily in hot water. It has 
but little taste. When heated, it loses water, and at 140 C. forms 
pyroboric acid (H 2 B 4 O 7 ). At a red heat, all the water is expelled, 
and boric anhydride (B 2 O 3 ) results; this is stable and non- volatile, 
even at high temperatures. Consequently, it will decompose nearly 
all metallic sulphates, carbonates, and nitrates when fused with 
them, forming metallic borates. Hence it is used as a flux. Boric 
acid is chiefly used in the preparation of borax; in enamels and 
glazes for pottery ; in making Guignet's green ; as an antiseptic in 
medicine and surgery ; and for preserving fish, meat, and milk. 

Borax, sodium biborate (Na 2 B 4 O7), is the only important salt de- 
rived from boric acid. It is found native in Thibet, Ceylon, and 
California. But little is known of the method of preparing borax 
in Thibet. It comes from that country as tinkal, an impure, crys- 
tallized borax, containing lime, magnesia, sulphates, and chlorides 
and a greasy substance added presumably to protect the crystal 
from efflorescence and breakage. The tinkal is purified by dissolv- 
ing in hot water, and adding lime-water and calcium chloride, to pre- 
cipitate the grease as lime soap. After filtering, the borax is crys- 
tallized by concentrating the solution. 

Borax was formerly obtained by evaporating the water and by 
washing the mud from the beds of several ponds (Borax Lake, and 
others) in Lake, Inyo, Kern, and San Bernardino counties in Califor- 
nia ; this was succeeded by its recovery from various dry lake beds, 
so-called " marshes," in the Death Valley region, where the surface 
efflorescence, or crusts, consists of a mixture of borax, soda, salt, and 
sulphate. At present nearly all borax produced in this country is made 

* F. Wittig (Zeit. angew. Chem., 1888, 483), recommends iron crystallizing 
tanks, because lead-lined vessels buckle and leak, owing to the changes of tempera- 
ture. The iron soon becomes polished, and yields perfectly clean crystals. 


from ulexite or " cotton ball " (NaCaB 5 O 9 8 H 2 O) and colemanite 
(Ca 2 B 6 On 5 H 2 O), found in Inyo, San Bernardino, and Ventura coun- 

The ore in small lumps is roasted at low heat in a rotary furnace, 
to expel water and cause the colemanite to fall to powder, which is 
sifted to remove the refuse calcite, sand, clay, etc. The powder is 
boiled in sodium carbonate and bicarbonate solution until decomposed, 
the calcium carbonate settled, and the brown solution of borax run 
into large iron crystallizing vats, which are sheathed with wood to 
prevent too rapid cooling. To form good crystals, the solution should 
cool very slowly, and the vats are usually covered to prevent the for- 
mation of surface crust. Wires are suspended in the vat for the borax 
to crystallize upon, which requires from 7 to 10 days. Impure crys- 
tals deposit on the bottom and sides of the vat, and are generally 
recrystallized from water. The sodium bicarbonate is added to 
prevent the formation of metaborate. The reactions are as follows : 

5 H 2 O + 2 Na 2 CO 3 + 2 NaHCO 3 = 

4 CaCO 3 + 3 Na 2 B 4 O 7 + 11 H 2 O. 
4 NaCaB 5 O 9 8 H 2 O + 2 Na 2 CO 3 + 2 NaHCO 3 = 

4 CaCO 3 + 5 Na 2 B 4 O 7 + 9 H 2 O. 

The sludge from the decomposing vat is boiled with water, filter- 
pressed, and the liquor sent to the next decomposing operation, while 
the mud is thrown away. 

A borocalcite called pandermite, found in Asia Minor and some 
other localities, is worked in somewhat similar manner. 

Much of the boric acid produced in Italy is converted to borax 
by boiling it with sodium carbonate ; the solution is concentrated to 
22 Be. at 104 C., settled and run into shallow, open, crystallizing 
vats, where the borax deposits within three days ; but for recrystal- 
lization deep tanks, tightly covered, and lagged to prevent radiation 
of heat, are used ; from 16 to 18 days are required, and large crystals 
are formed. 

Borax comes in trade in two forms : common or prismatic borax 
(Na 2 B 4 O 7 10 H 2 O), and octahedral borax (Na 2 B 4 O 7 5 H 2 O). The for- 
mer is produced as large, "efflorescent, monoclinic crystals, by crystal- 
lizing from a solution of 22 Be., which is permitted to cool to 27 C. ; 
it melts in its water of crystallization when heated, then swells greatly, 
forming a spongy mass, which fuses at red heat to a transparent glass. 
Octahedral borax is obtained as regular octahedrons, when a solution 


of common borax is concentrated to 30 Be., and cooled only to 
56 C. It is permanent in dry air, but absorbs moisture on exposure, 
and passes into the prismatic variety ; it fuses without intumescence, 
and is preferred as a flux for brazing and soldering. 

Borax is used as a flux in welding and brazing metals ; in enamel 
and glazes for metal ware and pottery ; in laundry work and in starch 
to increase the gloss ; in soaps, especially those intended for use in 
hard water ; for preserving meat ; as a mordant in dyeing ; for the 
ungumming of raw silk ; in medicine and pharmacy ; and with casein 
for the preparation of paste. 

Perborates,* derivatives of the acid HBO 3 , have become important 
industrially, as oxidizing agents. Sodium perborate (NaBO 3 4 H 2 O), 
is a stable, crystalline salt, produced from sodium peroxide and 
boric acid solutions ; in cold, aqueous solution it acts like hydrogen 


Hofmann's Bericht iiber die Entwickelung der Chemischen Industrie. 

1875, 324, 343. 

Handbuch der Kali-Industrie. E. Pfeiffer, Braunschweig, 1887. (Boracit.) 
Chemiker-Zeitung : 

1879, 46. (Boric Acid from Boracite.) 1887, 605. (Borax in Chili.) 
Third Annual Report of the California State Mineralogist, 1883. 
Die Stassfurter Kali-Industrie. G. Lierke, Wien, 1891. 
California State Mining Bureau, Bui. No. 24. The Saline Deposits of 

California, 1902. 
Zeitschrif t f iir angewandte Chemie : 

1888, 483. F. Witting. (Borax from Boronatrocalcite.) 1891, 367. 
1892, 241. Dr. Scheuer. (Boric Acid and Borax Industry.) 

Journal of the Society of Chemical Industry : 

1889, 857. C. N. Hake. (Borax Lake in California.) 1892, 683. 
Engineering and Mining Journal : 

(Borax.) 53, 8. 54, 247. 

* Compt rendu, 1904 (139), 796. J. Soc. Chem. Ind., 1904, 1145; 1905, 275, 
276, 332. 


The electric furnace * is used for one or more of the following rea- 
sons : (a) To secure temperatures higher than are attainable with 
combustion furnaces, thus making possible the production of certain 
substances previously unknown, or obtained with great difficulty; 
(6) for generating heat at the exact point required, thus avoiding 
the heat losses and depreciation of plant incident to forcing large 
quantities of heat through retaining walls; (c) for the maintenance 
of definite conditions (especially reducing atmosphere) within the 

The successful application of the electric furnace to technical 
uses by Messrs. Cowles, in Cleveland, Ohio, in 1884, was the begin- 
ning of large industries. Various modified forms of the Cowles fur- 
nace are now used to produce carborundum, artificial graphite, 
calcium carbide, phosphorus, alundum, barium hydrate and cyanide, 
and other products, and in metallurgical operations. 

The Cowles furnace (Fig. 89) consists of a crucible (F), into 
which the movable electrodes (E) pass. The cover has an opening 

(0) for the escape of the gases. The 
carbon electrodes are in contact at first, 
but are slowly separated as the charge 
and furnace become hot, and the cur- 
rent passes through the mixture in the 
crucible, or an arc is formed. At (J) 
the electrodes are joined to the conduc- 
tors from the dynamos. When the electrodes have been separated 
until the ammeter readings have become nearly constant, the opera- 
tion is allowed to go on for some hours. Either direct or alternating 
currents may be used, when the desired results can be obtained by a 
high temperature, and are not due to electrolysis. 

In some forms of electric furnaces the heating is accomplished by 
passing the current through a conductor of relatively high resistance 
embedded in the charge; the heat from the resistance warms the 
adjacent portions of the charge. 

Carborundum, or silicon carbide, was first made on a technical 

* Electric Furnaces and their Industrial Applications, J. Wright, New York, 
1905. The Electric Furnace, Alfred Stansfield, 2d Ed., New York, 1914. (Mc- 
Graw-Hill Co.) Trans. Faraday Society, Jan. 1905 (I), 85. 



scale by E. G. Acheson, about 1891, using the Cowles furnace. It is 
now extensively produced at Niagara and other places, and used as 
an abrasive, replacing emery and corundum. The charge of 100 
parts coke powder, 100 parts sand, and 25 parts common salt, to 
which a little sawdust is sometimes added, is packed around a hori- 
zontal core, twelve feet long, of granulated coke, joining the elec- 
trodes, which are embedded in the furnace walls. The heat causes 
the granulated coke to sinter together; the salt causes adhesion 
between the particles of the charge. As the reaction proceeds, 
large quantities of carbon monoxide are evolved, and the furnace is 
enveloped in blue flames. The reaction is : 

SiO 2 + 3 C = SiC + 2 CO. 

After several hours vapors of sodium appear and cause the flame to 
become yellow ; the furnace is permitted to cool, and the core is found 
surrounded by a crust of crystallized carborundum, with an inter- 
vening layer of graphite, formed by the decomposition of some of 
the carborundum by the heat. The brilliant black carborundum 
crystals often have a splendid iridescent lustre. The material is 
sorted by hand, and the carborundum crushed and washed with sul- 
phuric acid to remove traces of iron, aluminum, sulphides, phosphides, 
carbides, etc. It is then washed with water, and levigated to separate 
the powder into commercial sizes. 

Carborundum is not attacked by acids or by sulphur fumes, is 
stable in the air and infusible, and is harder than corundum. It is 
decomposed by fusion with caustic alkalies and nitre, and is attacked 
by chlorine above 600 C. 

Artificial graphite is made by heating amorphous carbon in the 
presence of ferric oxide or silica, at high temperature, so that the iron 
or silicon is vaporized and the carbon is left in the crystalline form 
as graphite. At intermediate temperatures in the presence of carbon, 
iron and silicon form carbides ; at higher temperatures, the carbides 
dissociate into the elements and the carbon formed by the dissociation 
is in the form of graphite. Thus iron and silicon act as catalyzers 
of the graphite formation. 

The brick furnace used is similar to the carborundum furnace 
and has carbon electrodes. Anthracite coal, as raw material, is filled 
into the furnace around a carbon rod as a core between the terminals, 
which heats the coal at the start since it is a poor conductor of heat 
when cold. Nearly all impurities are vaporized and the graphite 
contains only about 0.5 per cent of ash. This product is used for 



lubricators, paints, dry batteries, pencils, etc. Articles formed 
from pulverized amorphous carbon, pressed into moulds, can be 
" graphitized " in the electric furnace without change of form. 

Calcium carbide was first prepared on a commercial scale by 
T. L. Willson, about 1895, although it had been known as a labo- 
ratory product many years before. 

By heating an intimate mixture of pulverized lime and coke in 
an electric furnace, calcium carbide is formed directly : 

CaO + 3 C = CaC 2 + CO. 

The furnace (Fig. 90) * generally used is made of fire-brick and 
lined with carbon; it is designed for 3000 to 4000 kilowatts. The 
iron bottom of the furnace connects with the carbon lining of the 

FIG. 90. 

bottom, to form one electrode, and the other electrode is suspended 
so that it hangs free within the hearth. The fused carbide forms a 
pool under the electrode, which is raised or lowered as need be by the 
hoist (W). The furnace is tapped at intervals, by means of a special 
arc, -sprung at the end of the pointed, tapping electrode (A), by which 
a hole can be melted through the furnace wall in a few minutes. For 
a short time previous to tapping, no fresh charge is introduced and 
the fused carbide in the furnace forms a thin, liquid bath. The 
* Electrochem. Ind., 1908 (7), 400. 


charge itself serves to protect the fire-brick walls from the intense heat 
of the arc. 

The raw materials, which must not contain water, phosphate, sul- 
phate, nor mangesia, are lime or limestone, and coke, charcoal, or 
anthracite; these are coarsely pulverized and mixed in proportions 
of 95 kg. lime to 68 kg. coke. In theory, to produce 100 kg. of car- 
bide, 87.5 kg. of lime and 56.25 kg. of carbon are needed. One kilo- 
gram of carbide requires about 4 kilowatt-hours, and 1 ton of carbide 
is obtained from 1.79 tons of mixture of lime and coke. 

Calcium carbide is a hard, crystalline mass, with lustrous surface 
when freshly broken, but soon tarnishing and decomposing in the 
air. It reacts at once with water, forming acetylene and calcium hy- 
droxide : 

CaC 2 + 2 H 2 O = C 2 H 2 + Ca(OH) 2 . 

Commerical carbide contains about 80 per cent CaC 2 , and is chiefly 
used to prepare acetylene gas (p. 324), for the manufacture of calcium 
cyanamide (below), and somewhat as a germicide in combating 

Calcium cyanamid,* discovered by Franlc and Caro when attempt- 
ing the synthesis of cyanides (p. 290), is formed when purified and 
concentrated nitrogen gas (from liquid air) is brought into contact 
with finely ground calcium carbide, in ovens heated to about 1000 C. 
The reaction 

CaC 2 + N 2 ^ CaCN 2 + C 

is reversible if the conditions are not kept within certain limits, re- 
garded as trade secrets. 

The product from the ovens is a hard cake (black from the free 
carbon), with about 22 per cent nitrogen and 1 per cent carbide. 
After fine grinding and careful hydration of the residual carbide, the 
material goes to the fertilizer trade as " lime-nitrogen" or " nitrolim." 
It also finds use in making synthetic ammonia (p. 150) ; for cyanides 
by fusion with common salt ; and for case-hardening iron, especially 
for armor plate. 

Alundum is the name given to an artificial corundum (A^Oa), pro- 
duced by fusing bauxite in the electric arc furnace. Iron and most other 
impurities volatilize, leaving nearly pure aluminum oxide in the fur- 
nace. The cooled mass is pulverized in crushers and rolls, and sieved 
to the desired size of grain, for making into wheels and other imple- 

* Zeitsch. angew. Chem., 1903 (16), 536; 1910 (23), 2405. J. Soc. Chem. Ind., 
1903, 809. Electrochem. Met. Ind., 1907, 77 ; 1908, 341 ; 1910, 539 ; 1915, 213. 


ments for grinding and polishing, and for refractory linings and similar 
uses. As an abrasive it is harder and tougher than emery, which it 
has largely replaced. 

Barium hydroxide * is made in the electric furnace from barytes, 

thus : 

1) 4 BaSO 4 + 4 C = 3 BaSO 4 + BaS + 4 CO. 

2) 3 BaSO 4 + BaS = 4 BaO + 4 SO 2 . 

A mixture of ground barytes and coke, in the above proportions, 
is heated in an electric furnace which may be tapped periodically. 
The first reaction takes place at once and at moderate temperature, 
but the second is slower and requires very high heat. The product 
tapped from the furnace is dissolved in hot water, and the solution 
of hydroxide and sulphydrate filtered. Crystals of Ba(OH)2 8 H 2 O 
separate from the solution on cooling, the sulphydrate remaining in 
the mother-liquor. The crystals are centriffed, washed, and dried. 
The reduction of the barytes is claimed to equal nearly 97 per cent 
of the available sulphate, and the product is very pure. 

Cyanides J may be made in the electric furnace by heating a mix- 
ture of barium carbonate ^,nd coal or coke dust until barium carbide 
is formed, and then introducing nitrogen gas (deoxidized air), whereby 
barium cyanide is produced. The charge is cooled somewhat before 
the nitrogen is brought in contact with the mass. 

For electric carbon disulphide process, see p. 297. 

For electric phosphorus process, see p. 256. 


Applied Electrochemistry. By M. de Kay Thompson, New York, 1911. 
(Macmillan Co.) 

* J. Soc. Chem. Ind., 1902, 391. Trans. Am. Inst. Elec. Eng., 1902. 
f J. Soc. Chem. Ind., 1900, 745. U.S. Pat. Nos. 657,937 ; 657,938. 


Arsenious acid, white arsenic, or arsenic trioxide (A^Oa) is the 
most important arsenic derivative. It is made by roasting arsenical 
pyrites (mispickel), FeAsS; or as a by-product in the preparation of 
zaffre from cobaltite (CoAsS), or smaltite (CoAs 2 ), and in roasting 
certain arsenical tin ores before smelting. 

The roasting is done in reverberatory furnaces, and the vapors of 
white arsenic sublime off, and are condensed as a powder in long 
horizontal canals, or in chambers. The crude product is purified in 
a small reverberatory furnace, fired with coke, or in cast-iron pots, a 
number of which are set in a furnace, all being connected with a single 
condensing chamber or canal. Directly over the pot an iron drum or 
cylinder is often placed, from the top of which a short pipe leads to 
the condensing chamber. 

After resubliming, the oxide is a white granular powder, which 
is usually ground before packing for market; or, by a second subli- 
mation under slight pressure in an atmosphere of arsenious acid, it 
is obtained in an amorphous or vitreous state. For this the pot is 
heated red-hot, and the " arsenic meal " introduced through an open- 
ing in the cap of the drum, which is then closed. The arsenic vapor 
rises into the drum, and condenses on its walls as a transparent layer 
of " arsenic glass." 

White arsenic, or, as it is commonly called, arsenic, comes in com- 
merce as a powder, and as a " glass." On standing, the latter changes 
to a crystalline state, and becomes white, opaque, and porcelain-like 
in structure. It has no odor, and a very slight metallic taste, is diffi- 
cultly soluble in water, and vaporizes without melting when heated in 
the open air. It is used in glass-making ; when dissolved in glycerine, 
as a mordant in calico printing; in making various pigments; for 
preparing fly and rat poisons ; as a preservative for green hides ; for 
the manufacture of arsenic salts ; for insecticides ; in medicine ; and 
formerly, to a great extent, in the preparation of aniline from nitro- 

Arsenic acid, H 3 AsO 4 , is prepared by heating 4 parts arsenic tri- 
oxide with 3 parts concentrated nitric acid (1.35 sp. gr.), and evapo- 
rating the solution to a thick syrup, in which form it is usually sent 
to market. By evaporating it to dryness, and igniting at a red heat, 
arsenic pentoxide, A^Os, a hygroscopic body, is formed. 


Arsenic acid attacks the skin, producing blisters, but is less poison- 
ous than arsenious acid. It is chiefly used in calico printing, but was 
formerly much employed as an oxidizing agent in making certain coal- 
tar dyes (rosanilines). 

Sodium arsenate, Na2HAsO4, is made by heating white arsenic 
with sodium nitrate, or by dissolving white arsenic in sodium car- 
bonate solution, adding some sodium nitrate, evaporating to dryness, 
and calcining the mass. By dissolving in water, and crystallizing, the 
salt Na2HAsO4 12 H^O is obtained. This usually contains some 
NaH2AsO4 H 2 O (binarsenate). 

It is used as a substitute for the " dung-bath " in dyeing alizarines, 
and in calico printing, to prevent discoloration of the white parts of 
the pattern by rendering the excess of mordant insoluble, so that it 
does not " bleed," i.e. diffuse into the white portions of the cloth. 

Sodium arsenite, NaAsO2 (meta-arsenite), is prepared by neutral- 
izing arsenious acid with sodium carbonate, or hydroxide solution, 
and boiling for some time. The salt has been used instead of the 
" dung-bath " in dyeing. 

Orpiment and Realgar have been described on pp. 240 and 245. 

Lead arsenate,* PbHAsO 4 , or Pb 3 (AsO 4 )2, made by precipitation 
of a lead acetate or nitrate solution with sodium arsenate, is a white 
amorphous powder much used as an insecticide spray in agriculture. 
It is less injurious to foliage and adheres better than Paris green. 
It should not contain lead arsenite, which is more soluble and hence 
affects vegetation seriously. 

* Bui. No. 121, U. S. Bureau Chemistry, Dept. Agriculture, 1910. 


The substances sold under this name are silicates of sodium, or 
potassium, or of both. They are soluble in water, and are generally 
sold as thick, syrupy liquids. 

Commercial water-glass is not of definite composition, but is ap- 
proximately Na 2 Si4Og. It is prepared by fusing powdered quartz, 
or infusorial earth, with caustic soda or with sodium carbonate. A 
small quantity of charcoal is also added, to assist in the complete 
reduction of the carbonate. Sodium sulphate may be used instead of 
the carbonate. The fusion is done in a reverberatory furnace, and 
requires 8 or 10 hours. Sometimes ordinary glass-pots and furnaces 
(p. 199) are used. The product is a translucent or transparent glass, 
slightly green, from traces of iron. It is powdered, and boiled in 
water, best in a digester under pressure, until the soluble matter is 
dissolved. A small quantity of copper or lead oxide is added, to 
decompose any sodium sulphide formed during the reduction. After 
10 or 12 hours the solution is drawn from the boiler, filtered on cloth, 
and allowed to settle. It is then concentrated to 140 Tw. (1.7 sp. 
gr.). The material used must be pure, and especially be free from 
lime, alumina, etc. 

Water-glass is also made by boiling silica in a digester with a solu- 
tion of caustic soda for a long time at 60 pounds pressure. This 
yields a solution of the silicate directly, which needs only a little con- 
centrating. Sometimes gelatinous precipitated silica is dissolved in 
caustic soda, and the solution is evaporated. By using a mixture of 
equivalent weights of sodium and potassium carbonates, a more soluble 
glass is produced, which is sometimes called " double soluble glass." 

Potassium silicate, which forms a more soluble glass than sodium 
silicate does, is made in the same way. 

Water-glass is readily decomposed by acids, even carbon dioxide 
setting free silica, and forming a salt of the alkali. It is used exten- 
sively as an addition to yellow or laundry soaps ; as a fixative for 
pigments in calico printing ; as a vehicle for pigments in fresco paint- 
ing ; for rendering cloth and paper draperies non-inflammable ; as a 
size for paper and fabrics ; for preserving eggs ; as a preservative for 
timber and porous stone in the manufacture of artificial stone ; and 
in cement mixtures for glass, pottery, wood, and leather. 



Barium peroxide * BaO 2 , is made by calcining barium nitrate, and 
heating the oxide thus obtained in an atmosphere of dry, pure air. 
The nitrate is packed in crucibles, and heated in a furnace at 880 C. 
for several hours. The mass fuses, and for the first 3 or 4 hours 
continues to evolve nitrous gases, but finally becomes solid, though of 
a spongy, porous character. This is barium monoxide, and must 
be carefully protected from moisture and carbon dioxide. It is 
broken up into small lumps, and put into flat iron trays, which are 
set in wide, cast-iron pipes, thrpugh which a current of air can be 
passed. The air is dried thoroughly, and freed from carbon dioxide 
before it enters the pipes, by passing it through a drying tower, or 
drum, filled with caustic soda or quicklime. The pipes are heated to 
a low red heat (400 C.), and the air passes through them. The 
barium oxide takes up an atom of oxygen, forming the peroxide, 
while nitrogen escapes from the pipe. The product is cooled away 
from contact with air. 

By adding an excess of barium hydroxide solution to a solution 
of hydrogen peroxide, a precipitate of hydrated barium peroxide, 
BaO2 8 H 2 O is obtained, which is stable. By drying this at 130 C., 
all the water is expelled, and the pure peroxide remains. 

Barium peroxide is a gray or white powder, insoluble in water, 
but combining with it to form a hydrated compound. It is easily 
decomposed by dilute acids, and even takes up carbon dioxide from 
the air. Heated to a bright red heat (1000 C.), it decomposes into 
monoxide and free oxygen. Its chief uses are for making hydrogen 
peroxide, and in the preparation of oxygen gas. 

Hydrogen peroxide f H2O2, is made by decomposing barium perox- 
ide with dilute mineral acids. The finely powdered barium peroxide 
is actively stirred into diluted hydrochloric acid in which blocks of ice 
are floating. The temperature must not rise above 15 C. When all 
the peroxide is dissolved, dilute sulphuric acid in slight excess is added 
to precipitate the barium. Then to remove iron and alumina, some 
sodium phosphate is added, with more barium peroxide to make the 
solution neutral; finally add ammonia to decided alkaline reaction. 

* J. Soc. Chem. Ind., 1890, 246. L. T. Thome. Chemiker-Zeitung, 1894, 68. 
t Zeitschr. f. angew. Chem., 1890, 3. G. Lunge, J. Am. Chem. Soc., 12, 64. 
A. Bourgougnon. J. Soc. Chem. Ind., 1902, 229. 



The turbid liquor is rapidly put through a filter-press, and the clear 
filtrate immediately made slightly acid with sulphuric acid, as the al- 
kaline solution will not keep. Any barium remaining in solution is 
precipitated with pure sodium sulphate solution, and the liquor settled. 
Phosphoric acid may be used instead of hydrochloric and gives a 
stable product ; but it is more expensive. The commercial strength 
is known as a 12-volume solution, i.e. 3j per cent H 2 O 2 . 

By using hydrofluoric acid, the precipitate of barium fluoride 
may be readily employed to generate more of the acid ; if nitric acid 
is used, a considerable part of the barium is recovered as barium 
nitrate, with which more barium peroxide can be made. 

Hydrogen peroxide is a powerful oxidizing agent towards sub- 
stances capable of oxidation, but with bodies which give off oxygen 
readily it acts as a reducing agent, giving up one atom of oxygen to 
unite with the oxygen from the body in question, forming a molecule 
of the free gas. It is used extensively as a bleaching agent, especially 
for animal fibres and tissues, such as silk, wool, hair, feathers, bone, 
and ivory. It has long been used as a hair bleach for toilet use. As 
a disinfectant and antiseptic, it finds use in surgery; for restoring 
the colors of oil paintings which have darkened with age, it is very 
effective, if the paint contains lead; the lead sulphide is oxidized to 
the sulphate by the peroxide, the black color of the former being 
destroyed. Hydrogen peroxide has also been proposed as a substitute 
for sodium bisulphite and thiosulphate, as the reducing material for 
chrome tannage processes ; also as an antichlor, for use after chlorine 
bleaching; and as a general antiseptic, for use in the fermentation 
industries, and as a preservative for milk, beer, wine, and other fer- 
mentable liquids. 

Sodium peroxide,* Na2O 2 , has recently appeared in commerce as a 
bleaching material. The technical production depends upon the oxi- 
dation of fused metallic sodium, by exposing it to a current of pure 
dry air or oxygen. The sodium is contained in aluminum trays, 
which are put on cars, and pushed slowly through a wide iron pipe, 
externally heated to 300 C., while air, purified as described on 
p. 272, passes through the pipe in the opposite direction. The temper- 
ature must not rise above 300 C., and the oxidation must be slow. 

Sodium peroxide is a yellowish white, very hygroscopic powder, 
which is chiefly used as a powerful bleaching agent. It gives off 20 

* J. Soc. Chem. Ind., 1892, 1004 (Patent to H. Y. Castner) ; 1893, 603. Chem- 
ical Trade Journal, 11, 208. 


per cent of its weight as active oxygen.* It dissolves in dilute acids 
without evolving oxygen, if the vessel be kept cool, yielding a strong 
solution of hydrogen peroxide. It dissolves in water with the loss 
of some oxygen, and a great evolution of heat, which may be suffi- 
cient to set fire to inflammable bodies. It is too strongly alkaline 
for silk or wool bleaching, and should be converted into magnesium 
peroxide for this purpose. This is easily done by adding magnesium 
sulphate solution : 

Na 2 O 2 + MgSO 4 = Na 2 SO 4 + MgO 2 . 

The solution of sodium peroxide attacks cellulose, and produces 
an effect similar to that obtained by " mercerizing " with caustic soda. 

* Barium peroxide liberates 8 per cent of its weight of active oxygen, while a 
12-volume solution of hydrogen peroxide liberates only 1J per cent of its weight of 
active oxygen. 


Numerous processes have been devised for the technical produc- 
tion of oxygen, but most of them are so expensive, or require such 
complicated plants, that only two or three are in actual operation on 
a large scale at the present time. 

The decomposition of potassium chlorate by heating, with the 
addition of manganese dioxide, has been much employed, and is still 
the favorite laboratory method of obtaining a pure gas. The addi- 
tion of pyrolusite lowers the temperature of the decomposition, and 
reduces the liability of explosion. It is highly important that the 
potassium chlorate and pyrolusite be free from carbonaceous matter. 

Boussingault's process, as modified by Brin brothers, and worked 
on a large scale, is often called Erin's process.* Boussingault dis- 
covered that barium peroxide (BaO 2 ), when heated to a high tempera- 
ture, decomposes into the monoxide and oxygen, the latter passing 
off. Then by heating the barium oxide to a low red heat in a current 
of air, the peroxide can be regenerated. But his attempts to utilize 
the process were unsuccessful, because the monoxide soon became 
inert, and would not absorb oxygen from the air. This was due to 
the fact that the moisture and carbon dioxide in the air converted 
the barium oxide to hydroxide and carbonate, which are very stable 
bodies, even at high temperatures, consequently the regeneration of 
peroxide rapidly decreased. 

As modified by Brin brothers, the temperature of the retort re- 
mains constant, while all moisture and impurities are removed from 
the air. Barium oxide is made from barium nitrate, as described on 
p. 246, and put into vertical retorts, or long narrow pipes, suspended 
in a furnace heated by producer gas. When the temperature reaches 
700 C., purified air is forced into the retorts under a pressure of 15 
pounds per square inch, and the monoxide takes up an atom of oxygen, 
and forms the peroxide. The air supply is then cut off, and the 
pump reversed, so as to form a vacuum in the retort, reducing the 
pressure to about 26 to 28 inches of mercury. Under these condi- 
tions, the barium peroxide gives off an atom of oxygen, and is reduced 
to the monoxide. The gas is pumped into the gasometer, and when 
it ceases to be evolved the pump is reversed again, and air forced into 
the retort, to oxidize the monoxide to peroxide again. 

* J. Soc. Chem. Ind., 1890, 246. L. T. Thome. 1889, 82 and 517. 



The air is passed through purifiers, one filled with quicklime, and 
the other with caustic soda ; these remove the water and carbon di- 
oxide. By the alternate use of pressure and vacuum, the temperature 
may be kept constant at 700 C. The oxygen obtained is about 96 
per cent pure. The baryta is removed once in six or eight months, 
and broken up to prevent caking, after which it is returned to the 
retort. The yield of oxygen gas at each operation is said to be about 
10 litres per kilo of barium oxide employed. The cost of the gas in 
England is from 3s. to 7s. per 1000 cubic feet. 

Deville's process. By allowing sulphuric acid to drop in fine 
streams on red-hot surfaces, it breaks up according to the reaction : 

2 H 2 SO 4 = 2 S0 2 + 2 H 2 + O 2 . 

The gases evolved are passed through cooling coils to condense the water, 
and then through scrubbers containing water, to remove the sulphur 
dioxide. The retort is usually filled with broken brick, pumice, or other 
porous, acid-resisting material. The process has no significance as a 
method of preparing oxygen alone, but has been used for making sulphuric 
anhydride, SO 3 , the water being first condensed, and the sulphur dioxide 
and oxygen uniting to form the trioxide. About 114 litres of oxygen are 
obtained from 1 kilo of sulphuric acid. 

Tessie Du Motay process.* This depends on the following reac- 
tions : 

1) 2 Na 2 MnO 4 + 2 H 2 O = Mn 2 O 3 + 4 NaOH + 3 O. 

2) Mn 2 3 + 4 NaOH + 3 O(air) = 2 Na 2 MnO 4 + 2 H 2 0. 

First, sodium manganate is prepared by mixing a manganese oxide 
with caustic soda, and heating with free access of air. The following 
reaction takes place : 

2 Mn0 2 + 4 NaOH + O 2 (air) = 2 Na 2 Mn0 4 + 2 H 2 O. 

The sodium manganate is crushed, mixed to a paste with caustic 
soda solution, containing from 5 to 10 per cent NaOH ; this is dried 
slowly and completely in shallow pans, and ignited in a crucible at a white 
heat, to render it spongy. But it must not fuse. .This yields a porous 
manganate, containing an excess of caustic soda, which is filled into long 
clay or cast-iron retorts of peculiar construction, f set at an incline in 
the furnace, and heated to a regular temperature of 400-450 C. Super- 
heated steam is then admitted to the retort, where it deoxidizes the man- 
ganate, regenerating the manganic oxide and caustic soda, while oxygen 
is liberated, and is cooled and collected in a gasometer. Then the process 
is reversed, and purified air, which has passed through a heating pipe, 
set in the furnace, is admitted to the retort; it oxidizes the material, 

* J. Soc. Chem. Ind., 1892, 312. F. Fanta. 
t For details, see J. Soc. Chem. Ind., 1892, 315. 



regenerating the sodium manganate, while pure nitrogen escapes. The 
cycle of operations is repeated indefinitely. In order that the supply of 
oxygen may be continuous, the plant is usually built in duplicate, so 
that the contents of one set of retorts is being oxidized with air, while 
that of the other is being deoxidized with steam. 

The Linde refrigeration process* employs distillation and de- 
phlegmation of liquid air, which is made by the refrigerating effect 
produced when expanding compressed air from a higher to a lower 
pressure. At C., each decrease of one atmosphere pressure causes 
a drop of 0.276 C., in the temperature. The specific heat of a gas in- 
creases with increasing pressure, and the cooling effect is greater the 
lower the temperature at which expansion takes place. With suitable 
apparatus for heat interchange, the action of an indefinite number of 
expansions is accumulated and intensified, since the cold gas from 
each expansion serves to precool the compressed air before the next 
expansion. Air at 200 at- 
mospheres pressure enters 
the small copper tube (Fig. 
91) f and flows down 

through the triple Coil of Interchanger; 

the heat interchanger and 
finally through a copper 
coil submerged in the 
liquid air in the bottom hQ ' 91 ' 

of the rectifier. This lowers the temperature so much that the 
compressed air is liquefied before reaching the valve (A), by which 
the liquid is admitted to the top of the rectifying column which 
serves as a dephlegmator. Nitrogen, having a boiling point of 
195.5 C., tends to evaporate much faster than the oxygen boiling 
at 182.5 C. ; thus separation is effected in the column, oxygen 
descending as liquid, and nitrogen ascending as gas. The cold out- 
going gases pass through the heat interchanger coils surrounding the 
tube containing the compressed air, from which they absorb heat as 
they escape into the atmosphere. By allowing much of the liquid 
oxygen to vaporize also, a residual product of 95 to 98 per cent pure 
is obtained. 

The production of oxygen by the electrolysis of water (with some 
sodium hydroxide in solution) is practised commercially in this country 
and abroad. About 3 cubic feet of oxygen and 6 cubic feet of hydrogen 


* J. Soc. Chem. Ind., 1895, 984 ; 1903, 695. U. S. Consular Rep., 54, 64. 
t Publications of the Linde Air Products Company, Buffalo, N.Y. 


are obtained per kilowatt-hour ; each cell takes 350 amperes at 2 volts. 
The generator consists of an iron tank, about 3 feet diameter and 4 
feet deep, whose wall serves as the cathode. Suspended from the 
inside of the cover is a perforated steel cylinder, serving as anode ; an 
asbestos-cloth diaphragm surrounds the anode, separating it from the 
cell wall (cathode), and prevents mingling of the oxygen and hydro- 
gen, which pass off by separate pipes from their respective compart- 
ments. Both gases are obtained very pure. 

Oxygen is used for the oxy-hydrogen and oxy-acetylene flame, in 
melting platinum and other refractory metals ; for autogenous weld- 
ing and metal cutting ; in the calcium light ; in- purifying illuminat- 
ing gas; to destroy fusel oil in high wines; and in treatment of 
asphyxia and heart weakness. Its use has been proposed to hasten 
melting and refining of glass ; for enriching air in the blast-furnace 
and steel converter ; for oxidizing drying oils, and to assist the action 
of bleaching powder in textile work. 


Chemical Trade Journal, 1887, 145. 

Journal of the Society of Chemical Industry : 

1885,568. 1889,82,517. 1890,246. 1892,312. 1895,984. 1903,695. 

1911, 333. Ozone. 

Chemische Industrie, 1890, 104, 120; 1891, 71. G. Kassner. 
L'Ozone et ses Applications Industrielles. H. de la Caux. Paris, 1910. 


The sulphates of ammonium, magnesium, potassium, and sodium 
were discussed in connection with the industries to which they are 

Ferrous sulphate, green vitriol, or copperas, FeSO 4 7 H 2 O, is a 
by-product of several industries. Pyrites may be exposed to moist 
air until oxidation takes palce ; by lixiviation, a solution of ferrous 
and ferric sulphates, and sulphuric acid, is obtained, which is run over 
scrap iron. The reaction reduces all ferric salts, and the clarified 
and concentrated liquor yields light green crystals FeSO 4 7 H 2 O. 

FeS2 + H 2 O + 7 O = FeSO 4 + H 2 SO 4 . 

The basic ferric sulphate from the manufacture of aluminum 
sulphate from shale (p. 286) yields copperas by treatment with acid 
and scrap iron. The " sludge acid " of petroleum refining is some- 
times used for ferrous sulphate, by diluting and dissolving scrap 
iron in it. The acid " pickle liquors," used in foundries and wire 
mills for cleaning the surfaces of castings and wire, are treated 
with scrap iron to neutralize free acid, and yield more copperas 
on evaporation, than the market demands ; the disposal of the ex- 
cess is an industrial problem to prevent contamination of surface 

Wet metallurgical processes for producing cement copper (p. 616) 
furnish considerable copperas. Copper sulphide ores, low in copper, 
are weathered in heaps for several months, and frequently moistened 
with water. Oxidation of the sulphides forms copper and iron sul- 
phates, and when leached the liquors run into tanks containing scrap 
iron ; copper precipitates and ferric sulphate is reduced to the ferrous 
state. The solution is clarified and evaporated to crystallize. 

All processes for making ferrous sulphate yield dilute solutions, 
which are best evaporated by over-surface heating (p. 4), to prevent 
oxidation. The clarified liquid is put into lead-lined tanks, in which 
strings or wooden rods are suspended ; on these the large bluish green 
crystals of ferrous sulphate form. The crystals effloresce quickly 
when exposed to the air, and become coated with a brownish white 
powder of basic ferric sulphate, formed by oxidation ; ultimately 
the entire crystal is converted to this basic salt. By adding alcohol 



to a ferrous sulphate solution, the salt is precipitated in fine crystals 
which are more stable in the air than are the ordinary kind. 

Ferrous sulphate crystals have 7 molecules of crystal water ; when 
heated to 140 C., 6 molecules of water are expelled, but the last mole- 
cule is not removed until the temperature reaches 260 C., when basic 
salt begins to form. At a red heat, sulphuric anhydride is given off 
and ferric oxide is left. 

Copperas solution oxidizes quickly in the air, and a yellow pre- 
cipitate of basic ferric sulphate separates. Commercial green vitriol 
often contains copper sulphate, and sometimes nickel sulphate; if 
large quantities of these impurities are present, the color is very dark, 
and the salt is called " black vitriol." 

Ferrous sulphate is largely used as a mordant in dyeing; in the 
preparation of horticultural sprays; for disinfecting purposes; for 
the purification of water supplies ; in the manufacture of ink, Prussian 
blue, and various pigments ; and for precipitating gold from solution 
in metallurgical processes. 

Copper sulphate, blue vitriol, or "bluestone," CuSO 4 5 H 2 O, is 
now largely obtained as a by-product in the " parting " of gold and 
silver with sulphuric acid. The gold and silver alloy is boiled with 
concentrated sulphuric acid in cast-iron pans ; the silver is dissolved, 
the solution separated from the residue of gold, and the silver sulphate 
decomposed with metallic copper. Metallic silver precipitates, and 
copper sulphate remains in solution. 

Copper sulphate is also prepared by allowing sulphuric acid to 
drip on scrap copper with free access of air, the copper being slowly 
oxidized and dissolved. Or metallic copper, contained in lead-lined 
tanks, may be treated with hot acid. Scrap copper is often heated 
red-hot in a furnace, and then sulphur is thrown in, and the door 
tightly closed. Cuprous sulphide is formed, which is then oxidized 
at a red heat by admitting air into the furnace. A mixture of copper 
sulphate and oxide is thus produced, which is treated with hot dilute 
sulphuric acid, and the solution so obtained is evaporated. 

1) 2 Cu + S = Cu 2 S. 

2) Cu 2 S + 5 O = CuS0 4 + CuO. 

3) (CuSO 4 + CuO) + H 2 SO 4 = 2 CuSO 4 + H 2 O. 

Copper sulphide ores, chalcopyrite, and chalcocite, and artificial 
copper mattes are sometimes converted into blue vitriol; but the 
ferrous sulphate formed crystallizes with the copper sulphate. Such 


blue vitriol is much used where iron is not injurious. The iron may 
be removed by roasting the salt until the ferrous sulphate is decom- 
posed into oxide, and then dissolving in water and recrystallizing. 
Or the solution may be boiled with a little nitric acid or lead peroxide, 
until the iron is converted to the ferric state, when, by adding copper 
carbonate, or oxide, or barium carbonate, and boiling again, the iron 

Some copper ores contain zinc, and yield a bluestone, contami- 
nated with zinc sulphate. The acid " dipping liquors " from copper 
and brass works are also used for blue vitriol, but these are gener- 
ally contaminated with zinc. The hammer-scales (copper oxide), pro- 
duced in rolling and working sheet copper, are often dissolved in dilute 
acid to form blue vitriol. 

Copper sulphate forms deep blue crystals, containing 5 molecules 
of water. In dry air the crystals effloresce and fall to a white powder, 
but all the water does not escape until the mass is heated to 240 C. 
The anhydrous salt is a white powder, and will abstract water from 
alcohol or organic liquids. Bluestone is largely used as a mordant 
in calico printing, and in dyeing ; for preparing other copper salts and 
pigments ; in the preparation of germicides and insecticides (Bordeaux 
mixture, etc.), for batteries, and electrolytic baths; in metallurgy, 
and in most operations where a soluble copper salt is desired. 

Zinc sulphate or white vitriol, ZnSO 4 7 H 2 O, is not of very great 
importance. It is made by roasting zinc blende (sphalerite), or zinc- 
lead ores,* and leaching the mass with water or dilute sulphuric acid. 
Or scrap zinc is dissolved in dilute acid. The solution may be purified 
from copper by introducing a plate of metallic zinc, upon which the 
copper deposits. Iron is removed by heating the solution in the air 
for a considerable time, while stirring well, and then adding a small 
amount of zinc carbonate or oxide, to precipitate the ferric oxide. 

Zinc sulphate forms colorless crystals containing 7 molecules of 
water, which effloresce in the air. It is very soluble in water. When 
heated, the crystals melt in their water of crystallization, and at 100 
C., 6 molecules of water are expelled. The final molecule is driven 
off at 300 C., while at a red heat the anhydrous salt decomposes, 
leaving a residue of zinc oxide. 

Zinc sulphate is used somewhat in dyeing and printing ; as a 
disinfectant ; for preserving and clarifying glue solutions ; in medi- 
cine as an astringent, and in lotions; in the preparation of dryers 

* Bruno Kerl, Mineral Industry, 1895, 83. 


for " boiled oils " ; and to some extent as a preservative for hides 
and timber. 

Aluminum sulphate, A1 2 (SO 4 ) 3 , 18 H 2 O, is extensively employed in 
the arts, under the name " concentrated alum." It is usually pre- 
pared from pure kaolin, or from bauxite [Al2O(OH)4, or A^Os 2 H 2 O], 
or from the hydrated alumina obtained in the cryolite soda process 
(p. 113) or Bayer's process (p. 283). Aluminum hydroxide, prepared 
from bauxite or cryolite, is almost entirely free from iron, since it is 
precipitated from an alkaline solution of sodium aluminate, in which 
the iron of the mineral is not soluble. When this hydroxide is dis- 
solved in pure sulphuric acid, a very pure aluminum sulphate is 

(a). Aluminum sulphate from clay : China clay, free from cal- 
cium carbonate, is calcined at a moderate heat, until nearly all of its 
water is expelled; then it is powdered and sifted through very fine 
sieves, and mixed with a little less than the theoretical quantity of 
sulphuric acid of 1.45 to 1.50 sp. gr., and heated with free steam to 
start the reaction, which soon becomes very violent. The mass 
swells, and quantities of steam escape, but when the reaction ceases, 
the swelling subsides. If it is now allowed to cool, a stonelike sub- 
stance is obtained, which is employed in the arts as " alum cake." 
It contains all the silica and iron impurities of the clay, and usually 
from 2 to 3 per cent of free acid. But if the thick pasty mass is diluted 
with warm water while still hot, and decanted or filtered from the 
insoluble impurity, a solution of the sulphate is obtained, which on 
evaporation yields a salt containing about 0.2 per cent iron, and a 
trace of free acid. It is often customary to convert this solution di- 
rectly into alum (p. 285), by adding the necessary alkaline sulphate. 

(b). Aluminum sulphate from bauxite : Bauxite is more easily 
decomposed by acid than is clay, but if dissolved directly, the product 
contains a large amount of iron. However, considerable bauxite is 
decomposed with acid to form a hard cake which is known in trade 
as " alumino-ferric cake," and is used for many purposes where iron 
and free acid do no harm, and a cheap source of soluble alumina is 
desired, e.g. in precipitating sewage and waste liquors from dyeworks. 

But a pure sulphate is obtained by the following processes : The 
bauxite is roasted, powdered very fine, and mixed with calcined and 
finely powdered soda-ash, in the proportion of 1 molecule of Al 2 Os 
to 1.1 molecules of Na 2 O. If the bauxite contains much silica, 
more soda may be used, but the amount should not be sufficient to 


leave free carbonate in the product after calcination, otherwise the 
mass may fuse, and the solution of sodium aluminate obtained by 
lixiviating will be unstable. The mixture is calcined at a white heat, 
until all carbon dioxide and water are expelled ; this requires 3 or 4 
hours. The product is a porous, pale green or blue mass, which is 
ground and lixiviated with hot water, in a wooden tank, while stirring 
actively. A little caustic soda is added to the water, to prevent pre- 
cipitation of alumina (see Bayer's process, below). The lixivia- 
tion must be rapid, not occupying more than 10 minutes, after which 
the solution of aluminate is decanted. According to Jurisch,* the 
liquor should be at least 35 Be. density, and contain 170 grams A^Os, 
and 182 grams Na2O, per litre. Weaker solutions are said to yield a 
slimy precipitate of alumina, when decomposed in the next stage of 
the process. The liquor is quickly filtered (in a filter press), heated 
to 90 C., and decomposed by passing carbon dioxide into it, by which 
hydrated alumina is precipitated in a granular form, which is readily 
washed free from soda. The silica remains dissolved in the mother- 
liquor. The carbon dioxide may be derived from limekiln gases, 
or from the calcination of sodium bicarbonate. 

The pure aluminum hydroxide thus made is added slowly to hot, 
pure, concentrated sulphuric acid, until the frothing ceases ; the solu- 
tion, cooled in flat lead pans, forms a crystalline mass. If an excess 
of alumina is used in neutralizing, basic salt results. Sulphate made 
thus is nearly free from iron and silica, but may contain small quan- 
tities of soda. It is used in the arts under the name of " concentrated 
alum." From analysis, the formula appears to be Al^SOJs 20 H 2 O, 
but the excess of water may be hygroscopic and not combined. 

The process of J. K. Bayer f yields very pure alumina. A caustic 
soda liquor of 1.48 sp. gr. (47 Be.) is digested for six hours under 4 
atmospheres' pressure, at 170 C., with finely powdered bauxite, 
while actively stirred. The aluminum hydroxide of the bauxite dis- 
solves to form sodium aluminate solution, having about 1 A1 2 O3 to 
1.8 NazO. The solution is diluted to 1.20 sp. gr. (24 Be.), filter- 
pressed rapidly, and then decomposed in tanks by agitating for about 
72 hours, with a large excess of aluminum hydroxide. The hydroxide 
precipitates in crystalline form, until the proportions are about 1 
A1 2 O 3 to 6 Na 2 O ; silica and impurities remain in solution. A sufficient 
quantity of the milky liquid, carrying in suspension as much aluminum 

* Fabrikation von Schwefelsaure Thonerde, 52. 

t Jurisch, Ibid., 17-18. German patents, 43,977 (1887) and 65,604 (1892). 
J. Soc. of Chem. Ind., 1888, 625. 


hydroxide as was dissolved from the bauxite, is withdrawn and filter- 
pressed. The caustic soda liquor is again concentrated to 1.48 sp. 
gr. and the cycle repeated. The silica dissolved in the aluminate 
solution is precipitated during the digestion as an insoluble double 
silicate of sodium and aluminum (Na2Al 2 Si 3 Oio + 9 H 2 O), and remains 
with the residue, together with the iron. The hydrated alumina pre- 
cipitated is washed free from sodium salts, and dissolved in acid as 
described. It is also used for metallic aluminium (p. 638). 

Another process for sulphate consists in dissolving bauxite in 
dilute acid, at a temperature of 90 C., with the addition of a little 
sodium nitrate to oxidize all the iron to the ferric state; then more 
bauxite, together with a little potash alum, is added. After stirring 
thoroughly, the whole is left for several weeks. The iron combines 
with some of the alumina to form a precipitate: 

2 A1 2 (SO 4 ) 3 + 2 Fe(OH) 3 . 

(c). Sulphate from cryolite : The hydrated alumina obtained in 
the cryolite soda process (p. 113) may be dissolved to make aluminum 
sulphate in the usual way. The product may contain some soda. 

Another method of utilizing cryolite depends on the following re- 
actions : 

1) 6 NaF, 2 A1F 3 + 6 Ca(OH) 2 = 6 CaF 2 + 2 Al(NaO) 3 + 6 H 2 O. 

2) 2 Al(NaO) 3 + 6 NaF, 2 A1F 3 := 2 A1 2 O 3 + 12 NaF. 

3) A1 2 O 3 + 3 H 2 SO 4 = A1 2 (SO 4 ) 3 + 3 H 2 O. 

Powdered cryolite is boiled with milk of lime, and the solution 
of sodium aluminate decanted. By boiling the aluminate liquor for 
a long time, with more powdered cryolite, while stirring thoroughly, 
the second reaction takes place ; the residue is chiefly hydrated alu- 
minum oxide, while sodium fluoride goes into solution. By boiling 
the latter with milk of lime, caustic soda may be obtained as a by- 

2 NaF + Ca(OH) 2 = CaF 2 + 2 NaOH. 

By evaporating an aluminum sulphate solution until very con- 
centrated, and then cooling, a solid cake of the salt having a crystal- 
line structure is obtained ; its composition corresponds to 

A1 2 (SO 4 ) 3 20 H 2 O. 

It is difficult to obtain single crystals, but the usual formula assigned 
to them is A1 2 (SO 4 ) 3 18 H 2 O. The commercial product, however, 


never corresponds exactly to this formula. As now prepared, it con- 
tains but little free acid, or excess of alumina (basic salt), and only 
a minute trace of iron. It should contain 14 to 14.5 per cent Al 2 Oa, 
and dissolve readily in water to form a clear solution, i.e. no basic 
salt should be present. About 0.5 per cent free acid and 0.01 to 
0.1 per cent Fe 2 O3 are the average content of commercial samples. 
Since it may now be had of great purity, aluminum sulphate has 
largely replaced alum in the arts. It is extensively used as a mordant 
in dyeing; in preparing size for paper; for making alum and alu- 
minum salts (red liquor, etc.) ; in tawing skins ; for precipitating 
sewage or coloring matter from water; and, in general, for all pur- 
poses where alum was formerly used. 


An alum is a double sulphate of a univalent alkali metal and a 
hexad metallic radical of the form (R 2 ) == crystallized with 24 mole- 
cules of water. The general formula is therefore 

M 2 SO 4 , R2(SO 4 ) 3 24 H 2 O, 

or, as it is more frequently written, MR(SO 4 ) 2 12 H 2 O. The alkali 
metal may be sodium, potassium, ammonium, lithium, caesium, or 
rubidium. The hexad radical contains aluminum, chromium, iron, or 
manganese. In the majority of alums the essential part is aluminum 
sulphate, but since this does not crystallize well alone, it has, until 
recently, been difficult to obtain it pure enough for some purposes. 
But the addition of an alkali sulphate forms alum, which crystallizes 
beautifully and is very pure, while the alkali sulphate itself has no 
injurious action in most cases where aluminum sulphate is used. But 
since " concentrated alum " (p. 283) can now be had very pure, it is 
generally preferred, because of its greater strength and solubility. 

All alums crystallize, with the same number of molecules of water, 
in the regular system, either as octahedrons, or as cubes. They are 
all isomorphous, and a crystal of one kind of alum will continue to 
grow by accretion, if placed in a solution of another alum. Alum 
crystallizes from solution very perfectly, and forms exceedingly pure 
crystals, even from impure solutions. 

Alum occurs in nature in small quantities, produced by the action 
of volcanic gases on rocks consisting of potash-aluminum silicates; 
also in combination with iron and aluminum hydroxides in the mineral 
alunite, or alum stone, K 2 SO 4 , A1 2 (S0 4 ) 3 , 4 A1(OH) 3 , also formed by 


volcanic action. Other sources are alum slates and shales, clay, 
bauxite, and cryolite. 

Alunite, or alum stone, is insoluble in water. It is calcined in 
heaps, or in small shaft kilns, at about 500 C., and the mass is then 
exposed to the weather for several months, being moistened from 
time to time. The calcination converts the iron and aluminum hy- 
droxide into insoluble oxides, and the weathering forms alum in 
the mass, which is dissolved by lixiviation, and recrystallized. The 
alum thus obtained is basic, and crystallizes in cubes ; owing to im- 
perfect settling of the liquors before crystallization, some iron oxide 
is enclosed, giving the crystals a red color. This iron is, however, 
quite insoluble, and, no free acid being present, the alum yields a 
pure, neutral solution, and is especially desired for some purposes. 
It is made at Tolfa, near Rome, and so is called Roman alum. An 
imitation is made by coloring alum crystals derived in other ways, 
with brick dust, or with iron oxide (Venetian red). For further refer- 
ences on alunite see p. 158. 

Alum slates or shales are mixtures of iron pyrites, aluminum 
silicates, and bituminous matter. By exposure to the weather, the 
pyrites is oxidized to ferrous sulphate and sulphuric acid, and these 
react with the aluminum silicate to form aluminum sulphate. Basic 
ferric sulphate is also formed. The oxidation can be greatly hastened 
by roasting the shale before weathering it, but the temperature must 
not be high enough to drive off the sulphur. After weathering, the 
mass is systematically lixiviated, and a solution of aluminum sul- 
phate, having a specific gravity of about 1.16, containing some calcium 
and iron sulphates, comes from the leach tanks. This is clarified by 
settling, and some of the calcium and basic ferric sulphates deposit. 
The solution is evaporated in lead or iron pans by surface heating 
with direct flame, until ferrous sulphate crystallizes on cooling, and 
then the mother-liquor containing the aluminum sulphate is further 
concentrated to 1.40 sp. gr. During this evaporation, more calcium 
sulphate and a basic ferric sulphate separate. Scrap iron is generally 
placed in the vessel during concentration, to convert the ferric sul- 
phate into the basic salt, and to reduce the destructive action on the 
pan. The hot solution is decanted from the sediment, and mixed 
with potassium or ammonium sulphate in exact amount to form the 
alum. By agitating the liquid during the cooling, very fine crystals 
of alum, called " alum meal," separate. 

If the aluminum sulphate solution contains much iron, as is gen- 
erally the case when working on a large scale, it is often the practice 


to add potassium chloride to form the alum. By decomposing the 
iron sulphates, this forms potassium sulphate in the solution, and, at 
the same time, converts the iron into the very soluble ferric and 
ferrous chlorides, which remain in the solution when the alum sepa- 
rates. But with a pure solution of aluminum sulphate, this causes 
loss by converting part of the aluminum into the very soluble alu- 
minum chloride : 

4 A1 2 (SO 4 ) 3 + 6 KC1 = 3 5K 2 SO 4 , A1 2 (SO 4 ) 3 J + 2 A1C1 3 . 

The alum meal is washed with cold water in a centrifugal machine 
and recrystallized. It is sold both in the crystallized and in the 
powdered form. 

The manufacture of alum from clay, bauxite, or cryolite involves 
the preparation of a pure solution of aluminum sulphate by methods 
already given, and the addition of the exact quantity of alkali sulphate 
to form the alum. 

Blast furnace slag has been proposed as a source of alum. It is 
decomposed with hydrochloric acid, and the aluminum chloride solu- 
tion is decomposed with calcium carbonate ; the aluminum hydroxide 
so obtained is dissolved in sulphuric acid. The process is not suc- 
cessful, however. 

" Neutral alum " is made by adding sodium or potassium carbon- 
ate, or caustic soda to an alum solution, until a slight precipitate re- 
mains, even after vigorous agitation. After filtering, cubical crystals 
of the neutral alum can be obtained, but, as a rule, the neutral solu- 
tion is made by the user, and is not crystallized. Neutral alum is 
much used in mordanting, because of the great readiness with which 
it deposits alumina on the fibre. 

The most important alums of commerce are potassium alum, 
K 2 SO 4 A1 2 (SO 4 )3 24 H 2 O, and ammonium alum 

(NH 4 ) 2 S0 4 A1 2 (SO 4 ) 3 24 H 2 O. 

The latter is less soluble than the potash salt, but in all other respects 
they are quite similar. Both are stable in the air. 

Sodium alum, Na 2 SO 4 A1 2 (SO 4 ) 3 24 H 2 O, is very soluble in water 
and difficult to purify. Moreover, the crystals effloresce on exposure 
to the air ; in this condition, they are sometimes sold as " porous 

When heated, alum loses water and some sulphuric acid, and falls 
to a white powder, " burnt alum," which is difficultly soluble in water. 
This is used occasionally as a caustic in medicine. 


The chief uses of common alum are as a mordant in dyeing; in 
preparing size for paper-making; in tawing skins; in making pig- 
ment lakes; for clarifying turbid liquids, and precipitating sewage; 
and for hardening plaster of Paris casts, and other articles. 

Besides the common alums of trade, containing aluminum sul- 
phate as a basis, two others, iron alum and chrome alum, are also 
employed in the arts to some extent. 

Iron alum, which may be either (NH 4 ) 2 SO 4 , Fe 2 (SO 4 ) 3 24 H 2 O, or 
K 2 SO 4 , Fe 2 (804)3 * 24 H 2 O, is made by oxidizing a copperas solution 
to form ferric sulphate, adding the proper quantity of alkali sul- 
phate, and cooling below 10 C. It forms pale violet crystals, which 
are rather unstable, efflorescing and oxidizing in the air, forming basic 
ferric salt. Iron alum is chiefly used as a mordant. 

Chrome alum, K 2 SO 4 , Cr 2 (SO 4 ) 3 24 H 2 O, is largely produced as a 
by-product in the manufacture of alizarine. A mixture of potassium 
bichromate and sulphuric acid is employed to oxidize anthracene 
(CuHio) to anthraquinone (Ci 4 H 8 O 2 ), from which the alizarine is pro- 
duced. The effect of the reducing action of the organic body on the 
bichromate mixture is to form potassium and chromium sulphates in 
the solution in proper proportion to unite in chrome alum : 

CuHio + K 2 Cr 2 O 7 + 4 H 2 SO 4 = Ci 4 H 8 O 2 + K 2 SO 4 , Cr 2 (SO 4 ) 3 + 5 H 2 O. 

Chrome alum forms deep violet crystals, which effloresce on ex- 
posure to the air. It is used as a mordant; and in tawing skins, 
especially in certain chrome tannage processes. 


Die Fabrikation des Alauns, des Bleiweisses und des Bleizuckers. Dr. F. 

Junemann, Leipzig, 1882. (Hartleben.) 

Die Fabrikation von schwefelsaure Thonerde. K. W. Jurisch, Berlin, 1894. 
Journal of the Society of Chemical Industry : 

1882, 124. Newlands. 1883, 482. Kyiiaston. 1886, 16. Beveridge. 

1888, 625. (Bayer's Patent for Alumina Hydrate.) 1892, 4 and 321. 
Chemical News, 42, 191 and 202. 

Mineral Resources of the United States. 1893, 159 ; 1903, 265 ; 1904, 285. 
Bulletin No. 315, U. S. Geological Survey, 1906, 215. 


Cyanides are produced on a commercial scale by several methods, 
and a large number of patents which have been put into practice are 
described in chemical literature. Much energy and money have been 
expended in fruitless search for processes of cheap production of 
cyanides. In 1843 Langlois * showed that by passing ammonia gas 
over white-hot coke or charcoal, some ammonium cyanide is formed. 
This reaction has been the base of a patent to Lange and Emanuel,f 
in which the yield is improved by mixing hydrogen and nitrogen or 
deoxidized air with the ammonia : 

2 NH 3 + 2 C + 2 H = C 2 (NH 4 ) 2 . 
C 2 (NH 4 ) 2 + 2 N = 2 CN(NH 4 ). 

Cyanogen is present in crude coal gas and in some large gas works 
it is recovered by Bueb's process, t The gas is led direct from the 
tar extractor into a scrubber machine containing a ferrous sulphate 
solution. The hydrogen sulphide and ammonia in the gas react with 
the iron salt to form ferrous sulphide, which in turn precipitates the 
cyanogen as an insoluble salt of iron-ammonium cyanides; this is 
drawn from the machine as a black mud suspended in the liquor, and 
is filter-pressed. The reactions are : 

FeS0 4 + H 2 S + 2 NH 3 = 2 FeS + (NH 4 ) 2 SO 4 . 
2 FeS + 6 NH 3 + 6 CN + 3 H 2 O + 5 O = (NH 4 ) 2 Fe 2 (CN) 6 

+ 2(NH 4 ) 2 S0 4 . 

The solid cake is then decomposed with lime to form calcium 
ferrocyanide, which, in solution, is drawn off from the sludge and de- 
composed with potassium carbonate to yield potassium ferrocyanide. 
The ammonia is also recovered by distillation. If the ammonia is 
first removed from the crude gas by scrubbing, it is necessary to add 
alkali (Na 2 CO 3 ) to the copperas liquor in the cyanogen scrubber. 
Foulis process is based on this, a sodium ferrocyanide being formed. 

The recovery of cyanides from the spent oxide from the purifiers 
is described on page 29 1 

* Berzelius Jahresbericht, 22, 84. t German Pat. No. 122,280. 

t German Pat. No. 100,775. J. Soc. Chem. Ind., 1893, 511. 

U 289 


Bunsen and Playf air's process * for making cyanides by heating 
barium carbonate with powdered charcoal in an atmosphere of dry 
nitrogen was not a commercial success. It involved the reaction : 

BaCO 3 + 4 C + 2 N = Ba(CN) 2 + 3 CO. 

They also showed that the injection of heated air into a furnace 
containing carbon, alkaline earth oxides, and heavy metals produces 
cyanide ; thus the gases from blast furnaces contain these materials, 
and considerable attention has been given to recovering cyanides 
from the gases ; but as yet there has been no general introduction of 
these methods. 

Raschen's process t is based on the oxidation of sulphocyanide by 
means of nitric acid and atmospheric air. It is a continuous process, 
involving the following reactions : 

NaCNS + 2 HNO 3 = HCN + NaHSO 4 + 2 NO. 
2 NO + H 2 + 3 O = 2 HNO 3 . 

The apparatus consists of a series of earthenware jars, connected by 
earthenware pipes and so arranged that the liquor flows from near 
the middle of each jar, and passes to the bottom of the next. The 
gases from the decomposition contain prussic acid and much nitric 
oxide; they are scrubbed with water to remove the nitrogen 
oxides, and then the prussic acid is absorbed by caustic alkali 
and the solution evaporated in vacuum pans to prevent decom- 

Ammonium sulphocyanide (thiocyanate), NH 4 SCN, is sometimes 
prepared by Tscherniak and Giinzburg's modification of Gelis' pro- 
cess. | This depends on the following reactions : 

1) CSa + 2 NH 3 = NH 4 S 2 CNH 2 . (Ammonium dithiocarbamate.) 

2) NH 4 S2CNH 2 = NH 4 SCN + H 2 S. 

Carbon disulphide and ammonium hydroxide (0.91 sp. gr.), in 
proper proportion for reaction (1), are heated in an autoclave to 
125 C., while stirring actively. The steam is then cut off, but the 
stirring continued until the pressure rises to 15 atmospheres. This 
completes the first reaction, and the contents of the autoclave are 

* Rep. British Assoc., 1845. J. pr. Chem., 42 (1847), 397. 
t U. S. Pat. No. 567,552. Eng. Pat. No. 21,678 (1895). 
J Dingler's Polytechnisches Journal, 245, 214. 


blown off into a still,< which is heated to 110 C., at which point the 
ammonium dithiocarbamate is decomposed. The products of distil- 
lation are passed through condensers and scrubbers to collect volatile 
ammonium salts and carbon disulphide, while the hydrogen sulphide 
is conducted into a gasometer. The liquid in the still contains am- 
monium sulphocyanide, and is evaporated in tin vessels, and crystal- 

Sometimes lime and manganese peroxide are added to assist the 
reaction in the autoclave, in which case calcium sulphocyanide is 
formed : 

2 CS, + 2 NH 3 + MnO 2 + CaO = Ca(SCN) 2 + MnS + S + 3 H 2 O. 

Ammonium sulphocyanide and potassium ferrocyanide are now 
largely obtained from the spent iron oxide from the purification of 
illuminating gas. The spent oxide is first lixiviated with warm water 
(60 C.), until the liquor has a density of from 1.07 to 1.085. The 
solution, containing ammonium sulphocyanide and other ammonium 
salts, is evaporated to 1.2 sp. gr., and cooled, when the associated 
salts (ammonium sulphate, etc.) crystallize. The mother-liquor is 
further concentrated, and impure crystals of the sulphocyanide 
separate, which are purified by recrystallization. Ammonium sul- 
phocyanide is also obtained from gas-liquor by treating the non-vola- 
tile residue from the steam distillation (see Ammonia) with copper 
and iron sulphates, whereby cuprous sulphocyanide is formed. This 
is washed, and treated with ammonium sulphide, forming cuprous 
sulphide and ammonium sulphocyanide. The latter is then extracted 
with water. 

Ammonium sulphocyanide is very soluble in water and in alcohol. 
It is used as a source of other sulphocyanides, and in dyeing, to pre- 
vent the injurious action of iron on the color. 

The residue from the lixiviation is mixed with quicklime (which 
is slaked by the moisture in the damp mass), and heated by steam 
in closed vessels to 100 C. The lime decomposes the ferric ferrocy- 
anide and the double iron-ammonium cyanides, setting free ammonia 
gas, which is absorbed in scrubbers, and forming calcium ferrocyanide, 
which is obtained by lixiviating the mass. The solution of calcium 
ferrocyanide is evaporated, and treated with the calculated amount 
of potassium chloride to form the difficultly soluble calcium-potassium 
ferrocyanide, CaK 2 Fe(CN) 6 . This is separated from the mother- 
liquor, washed, and decomposed with potassium carbonate to form 
potassium ferrocyanide. 


The reactions are : 

1) Fe 4 SFe(CN) 6 J , + 6 Ca(OH) 2 = 3 Ca 2 Fe(CN) 6 + 4 Fe(OH) 3 . 

2) (NH4) 3 Fe 3 SFe(CN) 6 J , +'6 Ca(OH) 2 = 

3 C a2 Fe(CN) 6 + 3 Fe(OH) 3 + 3 NHs + 3 H 2 O. 

3) C a2 Fe(CN) 6 + 2 KC1 = CaK2Fe(CN) 6 + CaCl 2 . 

4) CaK 2 Fe(CN) 6 + K 2 CO 3 = K4Fe(CN) 6 + CaCO 3 . 

Potassium ferrocyanide, K4Fe(CN) 6 3 H 2 O, also called yellow 
prussiate of potash, is made by fusing together potassium carbon- 
ate, iron borings, and nitrogenous organic matter of any kind (horn, 
hair, blood, wool waste, and leather scraps).* The potash is fused 
in a shallow cast-iron pan, set in a reverberatory furnace, and the 
organic matter, mixed with from 6 to 8 per cent of iron borings, 
is stirred in, in small portions at a time, until about Ij parts of the 
mixture for each part of potash have been added. The temperature 
must be kept high enough to keep the mass perfectly liquid, but not 
hot enough to volatilize the cyanogen salts. The reaction is violent 
at first, and when the liquid remains in quiet fusion the process is 
ended, and the melt is ladled into iron pans to cool. The mass, con- 
taining a number of substances (KCN, K 2 CO 3 , K 2 S, FeS, metallic 
iron, carbon, etc.), is broken up into lumps the size of an egg, and 
digested with water at 85 C. for several hours. During this process 
reactions take place between the potassium cyanide and iron sulphide, 
by which the ferrocyanide is formed : 

6 KCN + FeS = K 2 S + IQFetCNV 

Liebig explained the reactions during the fusion as follows : part 
of the carbon and nitrogen of the organic matter combine to form 
cyanogen (CN) 2 , while some of the potash is reduced by the excess of 
carbon to metallic potassium, which at once unites with the cyanogen 
to form potassium cyanide. The sulphur in the organic matter com- 
bines with the iron, forming ferrous sulphide. Finally, on lixiviating, 
the formation of the ferrocyanide takes place. The solution is evapo- 
rated in iron pans by the waste heat of the furnace, and clarified while 
hot ; on cooling, the crude ferrocyanide crystallizes, and is purified by 
recrystallization. The mother-liquors yield more impure salt on 
further evaporation. 

* The organic refuse is sometimes partially charred in retorts, by which much 
ammonia is driven off and saved. But the yield of ferrocyanide is then less, since 
the nitrogen content of the char is small. 


The calcium ferrocyanide liquor from gas purification (p. 289) 
yields potassium ferrocyanide by treatment with potassium carbonate, 
filtering, and evaporation to crystallization. 

Potassium ferrocyanide forms splendid large lemon-yellow crys- 
tals, having 3 molecules of crystal water, which it gives off at 100 C., 
and is converted to a white powder. It is not poisonous. It is 
largely used for making Prussian blue ; in calico printing, and in 
dyeing ; for case-hardening iron ; for making potassium cyanide and 
ferricyanide ; and to a small extent in explosives, and as a chemical 

Barium sulphocyanide, Ba(SCN) 2 , is made by heating ammonium 
sulphocyanide with barium hydroxide solution, under slight pressure. 
Ammonia distils off, and the liquid is evaporated to yield the barium 
salt, Ba(SCN) 2 2 H 2 O. This is generally used for making potassium 
and aluminum sulphocyanides, KSCN and A1(CSN), which are used 
in textile dyeing and printing. 

Potassium ferricyanide, red prussiate of potash, K 3 Fe(CN) 6 , is 
usually made by passing chlorine gas into a solution of the ferro- 
cyanide, until ferric chloride no longer forms a precipitate, only 
producing a brown color in the liquid. It may also be made by 
exposing the dry powdered ferrocyanide to chlorine until a test 
portion, dissolved in water, gives nothing but a brown color with 
ferric chloride. 

2 K4Fe(CN) 6 + 2 Cl = 2 KC1 + 2 K3Fe(CN) 6 . 

Excess of chlorine must be avoided, since this forms a dirty green 
precipitate (Berlin green) in the solution, which cannot be removed by 

Lunge * recommends boiling the solution of ferrocyanide with 
lead peroxide, while passing a stream of carbon dioxide through the 
liquor : 

2 K4Fe(CN) 6 + H 2 O + O = 2 K3Fe(CN) 6 + 2 KOH ; 

but the final reaction may be written : 
2 K4Fe(CN) 6 + PbO 2 + 2 CO 2 = 2 K3Fe(CN) 6 + PbCO 3 + K 2 CO 3 . 

An excess of carbon dioxide is necessary to prevent decomposi- 
tion of the ferricyanide by the lead oxide and alkali. 

A very good product is obtained by the action of potassium per- 

* Dingier' s Polytechnisches Journal, 238, 75. 


manganate on a mixture of calcium and potassium ferrocyanide solu- 
tions : 

3 Ca 2 Fe(CN) 6 + 7 K 4 Fe(CN) 6 + 2 KMnO 4 = 

10 K 3 Fe(CN) 6 + 6 CaO + 2 MnO. 

The calcium and manganese hydroxides formed are removed from 
the solution by carbon dioxide, and the ferricyanide purified by 

Recently, anodic oxidation of a ferrocyanide solution, to form 
the ferricyanide, has been introduced. 

Potassium ferricyanide crystallizes in blood-red prisms, without 
crystal water, and is very soluble, forming a solution of an intense 
yellow color. With ferrous salts, it gives the blue pigment, Turn- 
bull's blue. With ferric salt, it gives a brown coloration, but no pre- 
cipitate. Its solution, with caustic potash, is a powerful oxidizing 
liquid, and as such is used in calico printing for a " discharge " on 
indigo and other dyes. It also forms part of the sensitive coating 
for " blue print " papers. It has been recommended for use with the 
potassium cyanide solution in gold extraction. 

Potassium cyanide, KCN, is generally made by fusing the ferro- 
cyanide with potassium carbonate, until the evolution of gas ceases. 
The following is the reaction : 

K 4 Fe(CN) 6 + K 2 CO 3 = 5 KCN + KCNO + CO 2 + Fe. 

The metallic iron separated sinks to the bottom of the crucible, and 
the fused mixture of cyanide and cyanate is run off. The addition 
of powdered charcoal reduces part of the cyanate to cyanide. The 
product is pure enough for many purposes. The cyanate, which is 
sometimes injurious, may be reduced by the action of metallic zinc 
or sodium, or the cyanide may be extracted with alcohol, acetone, or 
carbon disulphide. 

By fusing the ferrocyanide with metallic sodium, a mixture of 
sodium and potassium cyanides is obtained, which is extensively 
employed in the arts as " potassium cyanide." The so-called " cyan- 
salt " is made by fusing the ferrocyanide with sodium carbonate ; 
this is cheaper than the pure potassium salt. 

Potassium cyanide is also made by fusing the dry ferrocyanide 
in closed crucibles, until nitrogen ceases to be given off. Carbide of 
iron is formed, and sinks to the bottom of the crucible, if the fusion 
is allowed to stand for a considerable time. But the separation is 
imperfect, and the product is usually dissolved in alcohol or acetone, 


and the clarified solution heated in a still to recover the solvent. 
The product is then heated until it fuses, and when cold, it forms a 
white, transparent mass. Air must be carefully excluded during 
the whole process, to prevent the formation of cyanate. The re- 
action is 

K 4 Fe(CN) 6 = 4 KCN + FeC 2 + N,. 

But the product is not entirely free from potassium carbonate, 
since it is practically impossible to evaporate a cyanide solution 
without some decomposition and escape of the weak hydrocyanic 
acid. The caustic potash thus formed then combines with carbon 
dioxide from the air. Water cannot be used to leach the iron car- 
bide residue, since the potassium cyanide in solution at once recom- 
bines with the iron to form ferrocyanide again. 

Potassium cyanide is made from the sulphocyanide, by extract- 
ing the sulphur with zinc or lead.* The zinc is melted in a graphite 
vessel, and charcoal powder is spread over its surface. The sulpho- 
cyanide is stirred into the fused metal until the mass becomes a 
thfck paste, when it is allowed to cool. It is then systematically 
lixiviated in tanks similar to Shank's apparatus (p. 97.) Any alkali 
sulphide is precipitated by adding lead cyanide. The solution is 
evaporated in vacuum, and yields an impure product, containing 
cyanate and double zinc-potassium cyanide. 

Beilby's process f consists in passing dry ammonia gas through a 
fused mixture of potassium carbonate and carbon. A little potas- 
sium cyanide is added to increase the fusibility of the charge. The 
process is conducted in a covered cast-iron pot, or in a vertical retort 
having revolving rakes to stir the charge, and the fumes pass to a 
dust chamber. When the desired percentage of potassium cyanide 
has been reached in the fused mass, the charge is tapped off through 
a strainer to retain suspended carbon and run direct into drums. A 
similar method by Siepermann is worked in Germany. 

Castner's process involves the passing of dry ammonia gas over 
metallic sodium at a temperature of 350 C., and immediately run- 
ning the sodamide thus formed through layers of red-hot charcoal ; 
or a fusion of sodium cyanide and metallic sodium is mixed with 
powdered charcoal, and ammonia is passed through it. 

NH 3 + Na = NaNH 2 + H. 
NaNH 2 + C = NaCN + 2 H. 

* J. Soc. Chem. Ind., 1892, 14. 

f Ibid., 1892, 747, 1004. Eng. Pat. No. 4820, 1891. 


Potassium cyanide comes in commerce as white lumps or powder, 
very soluble in water and having alkaline reaction. It smells some- 
what like bitter almond oil, owing to the prussic acid liberated from 
it by the action of carbon dioxide and moisture in the air. On stand- 
ing, or when warmed, its aqueous solution decomposes, yielding am- 
monia and potassium formate : KCN + 2 H 2 O = NH 3 + HCOOK. 
When heated with reducible substances, it has strong reducing prop- 
erties ; hence its use as a flux in assaying and metallurgy. It is ex- 
tensively used in electroplating solutions, forming soluble double 
cyanides with gold, silver, copper, and other metals, in which the 
metal-ion concentration is very small, thus giving favorable condi- 
tions for a good deposit. Its largest use is for the recovery of gold 
from low-grade ores, and tailings of other reduction processes (p. 
631). A weak solution is used to dissolve the gold, forming aurous 
potassium cyanide, AuCN KCN. Formerly it was employed in 
photography to " fix " the image of negatives and prints, but has 
now been displaced by sodium thiosulphate (" hypo "). Potassium 
cyanide is extremely poisonous, both when taken internally and when 
introduced into the blood directly. 

The commercial salt usually contains cyanate and carbonate, 
and is sold in several grades ; the pure potassium salt contains about 
40 per cent of cyanogen, while sodium cyanide contains about 53 
per cent cyanogen, thus an impure potassium cyanide containing 
sodium cyanide may, by analysis based on the cyanogen content, 
appear to be 100 per cent pure, or even higher, if estimated as KCN. 
Commercial grades may assay as low as 65 per cent, but 95 to 98 per 
cent is customary. 


The Cyanide Industry. R. Robine and M. Lenglen. Trans, by J. A. 
LeClerc, New York, 1906. 

Coal Gas Residuals. Frederick Wagner, New York, 1914. (McGraw- 
Hill Co.) 


Carbon disulphide, 82, may be made by passing sulphur vapor 
over coke or charcoal, at a red heat (higher temperatures are not 
necessary). This was formerly done in iron or fire-clay retorts * 
heated from without, but destruction of the retorts was rapid. An 
improved apparatus (Fig. 92), devised by 
Taylor, f makes use of electrical heating, which 
localizes the heat within the retort and makes 
it possible to keep the walls relatively cool, 
thus decreasing the wear and tear. Sulphur 
is put into the chamber (Z) and partly sur- 
rounds the carbon electrodes (E). Fragments 
of coke (J ) fill the space between the electrodes 
and are fed to the furnace through (K, K), 
thus maintaining the continuity of electrodes. 
The shaft of the furnace is filled through (X) 
with charcoal t (Y). Crushed sulphur is fed 
through (V, V) and (R), filling the chambers 
(0) and (U). An alternating current is ap- 
plied through the electrodes, the sulphur in 
(Z) melts, and rising around the electrodes 
cuts off the contact more or less, and the fur- 
nace is partly self -regulating. The heat zone is at the top of the 
melted sulphur layer, and the vapor rises through the charcoal (Y), 
which has become sufficiently hot to form carbon disulphide, the 
vapor passing through (P) to the condensers. The furnaces are 41 
feet high by 16 feet in diameter. 

The crude carbon disulphide is impure and has a very offensive 
odor. It is purified to remove hydrogen sulphide, free sulphur, etc., 
by redistillation in a steam-heated still with a little caustic soda, or 
anhydrous copper sulphate, in the still ; or by washing with lime- 
water, followed by redistillation over a solution of lead acetate. 

Carbon disulphide is a pale yellow, or colorless, heavy, mobile 
liquid, having a fetid odor when impure, boiling at 46 C., and ex- 

FIG. 92. 

* J. Soc. Chem. Ind., 1889, 93. 

t Trans. Am. Electro. Chem. Soc., 1 (1902), 115 ; 2 (1902), 185. 
Ind., 1902, 353, 979, 1236. 

J Necessary to secure rapid reaction with the sulphur. 


J. Soc. Chem. 


tremely volatile at ordinary temperatures. Its vapors inflame at 
149 C., are very heavy, and are poisonous when breathed. It is sent 
to market in sheet iron cans, or drums, and is regarded as dangerous 
freight because of its extreme volatility, and the explosive nature of 
its vapor when mixed with air. When burned, it produces large quan- 
tities of suffocating gases (CO2, 802). It is only slightly soluble in 
water, but mixes well in all proportions with ether, benzene, alcohol, 
and many oils. It dissolves sulphur, phosphorus, iodine, camphor, 
wax, tar, resins, rubber, and nearly all oils and fats. Hence its use 
as a solvent and extractive agent is extensive. It is also used as a 
disinfectant; as a germicide and insecticide in agriculture, and in 
museums and herbariums ; in refrigerating machines ; for exterminat- 
ing moles, rats, woodchucks, and other burrowing animals ; in the 
manufacture of rubber cement ; in making cyanides and carbon tetra- 
chloride ; and in organic preparation work. 


Carbon tetrachloride is made by passing a mixture of carbon di- 
sulphide vapor and chlorine through a red-hot porcelain tube.* A 
mixture of sulphur chloride, S2C12, and carbon tetrachloride results, 
which is treated with milk of lime, and digested with potash, and the 
tetrachloride distilled. Or dry chlorine may be led into carbon 
disulphide containing a little iodine in solution, f The tetrachloride 
is distilled off, and washed with alkali, to remove iodine and sulphur 


+ 6 Cl = CC1 4 + S2C1 2 . 

Carbon tetrachloride is a heavy, colorless liquid, boiling at 76 C. 
It is a good solvent for many substances, and may be used instead of 
chloroform or carbon disulphide for extractions and is less poisonous 
than the latter. At temperatures but little above its boiling-point, 
it dissociates and hydrolyzes in the presence of water, forming 
chlorine and hydrochloric acid. This limits its uses as a solvent. 
It is not inflammable and is used in some types of fire-extinguishers. 

* Kolbe, Annalen der Chemie und Pharmacie, 45, 41 ; 54, 145. 
t Lever and Scott, English Patent No. 18,990, 1889. 


Sodium manganate, Na2MnO4, is made by mixing sodium nitrate 
or caustic soda solution with powdered pyrolusite, or manganese 
oxides, evaporating to dryness, and calcining the mass at a red heat, 
with access of air, in shallow vessels. The following is the reaction 
involved : 

MnO 2 + 2 NaOH + O = Na 2 MnO 4 + H 2 O. 

The product of the fusion is a dull green, porous mass, which, if 
lixiviated, yields a green solution of the manganate. But this is 
unstable, and if exposed to the air, or treated with an acid, or boiled, 
the manganate is converted into permanganate : 

3 Na 2 MnO 4 + 2 H 2 O = 2 NaMnO 4 + 4 NaOH + MnO 2 . 

In alkaline solution, however, the manganate is more stable. 

Sodium manganate is a powerful oxidizing agent, and is used as a 
disinfectant. It is also converted to the permanganate, and sold in 
solution as " Condy's liquid " for disinfecting purposes. Sodium per- 
manganate does not crystallize well. 

Potassium manganate, K 2 MnO 4 , is very similar to the sodium salt, 
and is made in the same way. It is chiefly used in preparing the 
permanganate, KMnO 4 , which crystallizes very well. This, being 
easily purified, and stable when crystallized, is the most important 
permanganate of commerce. It was formerly made by decomposing 
potassium manganate with sulphuric acid, carbon dioxide, or chlorine, 
followed by recrystallization. 

3 K 2 MnO 4 + 2 H 2 SO 4 = 2 KMnO 4 + 2 K 2 SO 4 + MnO 2 + 2 H 2 O. 
3 K 2 MnO 4 + 2 CO 2 = 2 KMnO 4 + 2 K 2 CO 3 + MnO 2 . 
2 K 2 MnO 4 + C1 2 = 2 KMnO 4 + 2 KC1. 

It is now made by anodic oxidation of the manganate (made as above) 
in alkaline solution, the cell having a porous diaphragm : 

2 K 2 Mn0 4 + O + H 2 = 2 KOH + 2 KMnO 4 . 

The permanganate crystallizes and settles to the bottom of the anode 
compartment, from which it is " fished " out at intervals. The caustic 
potash formed migrates to the cathode, whence it is continuously 



removed and returned to the manganate fusion. By this process no 
foreign substances are introduced, and the conversion of the man- 
ganate to permanganate is complete. 

Potassium permanganate forms deep purple, prismatic crystals, 
which dissolve in 16 parts of cold water. The solution has a power- 
ful oxidizing action, and can only be filtered on glass-wool or asbestos. 
When mixed with organic matter, the dry powder is subject to spon- 
taneous combustion, and forms explosive mixtures with easily oxidiz- 
able substances. It is used as a disinfectant ; in bleaching and dye- 
ing ; for coloring wood a deep brown ; for purifying ammonia and 
carbon dioxide gases ; and in medicine. 



WOOD consists mainly of cellulose (CeHioOs),^ with its incrusting 
layer of lignin, and of sap, containing water, resins, tannins, coloring 
matter, and mineral salts. Air-dried wood contains 15 to 20 per 
cent of moisture. When heated in closed retorts, away from the air, 
the cellulose and ligneous matter decompose, after the moisture is 
expelled, and a complex series of reactions occurs,* by which a great 
number of substances are formed. The crude products are gases, 
thin liquids, viscous liquids or tar, and charcoal. When wood is car- 
bonized in pits (p. 35), the volatile products go to waste ; by the use 
of retorts, the valuable liquid distillates and tar are saved; the 
gases evolved are mainly hydrogen, methane, ethane, ethylene, carbon 
monoxide, and carbon dioxide ; they have no value for illuminating, 
and are burned under the retorts, thus economizing fuel. 

When wood is heated in retorts, the moisture is driven out, but 
no decomposition occurs until the temperature approaches 160 C. ; 
between 160 and 175 C., a thin, watery distillate, called "pyroligneous 
acid," is formed; above 275 C., the yield of gaseous products be- 
comes marked, and between 350 and 450 C., liquid and solid hydro- 
carbons are principally formed. Above this last temperature, little 
change occurs, and charcoal, containing the mineral ash, remains in 
the retort. 

The pyroligneous acid contains the important distillates, methyl 
alcohol and acetic acid, together with acetone, methyl acetate, phenols, 
ketones, and other substances. The tar contains aromatic hydro- 
carbons and paraffines. Its most valuable constituent is the creosote 
oil, containing guaiacol, creosol, and other phenols of high molecular 

* Zeit. angew. Chem., 1909 (22), 1205. 



weight. A comparatively small amount of phenol or carbolic acid is 
present, however. 

The proportion of gaseous products to liquid distillate and char- 
coal is affected by the method of heating; rapid heating to a high 
temperature increases the quantity of gas; by distillation at a low 
temperature, the yield of pyroligneous acid, tar, and charcoal is larger. 
The variety of wood used affects the amount of acid and tar ; decidu- 
ous trees, especially birch, oak, and beech, are preferred ; coniferous 
woods yield less acid, but afford a tar containing much resin and 
turpentine. The yield of acid and tar is increased by the rapid re- 
moval of the vapors from the retort. 

Wood is distilled in various kinds of kilns or retorts. If charcoal 
is the only product in view, the carbonization may be done in " pits " 
or kilns (p. 35), and the volatile products go to waste. Masonry 
kilns of large capacity (15 to 90 cords at a charge) are often used ; 
the necessary heat may be derived from combustion of part of the 
charge itself, or from an external fire whose combustion gases pass 
into the kiln, or through flues in its walls. 

When the volatile products are to be saved boiler-plate iron retorts, 
externally heated, are employed; these may be either stationary or 

movable, and of large or small 
capacity. Small horizontal cyl- 
inders (Fig. 93,* A), holding 
about 1 or 1 J cords at a charge, 
are usually set two in a bench, 
over a common furnace. Each 
retort connects with a separate 
condenser (C) by the copper 
pipe (B) ; a pipe (D) carries the 
uncondensed gases to the grate where they are burned under the re- 
torts. The wood, cut to proper length, is rapidly filled into the retort, 
which is still hot from the previous charge, the door is closed and 
luted and distillation begins at once. After some 12 hours, the hot 
charcoal is rapidly drawn into an iron box to cool out of contact 
with the air, and the retort at once recharged. 

For larger output, horizontal oven-retorts (Fig. 94) are used. These 
are rectangular iron boxes, into which several steel cars, each loaded 
with 2 to 4 cords of wood, are run at one time, and the carbonization 
carried on for 24 hours. Thus charges of 10 to 20 cords of wood at 
a time are expeditiously handled. When carbonization is finished, 

* J. Soc. Chem. Ind., 1897, 667 and 722 (M. Klar). 

FIG. 93. 



the doors at each end of the retort are opened, and a string of newly 
loaded cars pushed in, which also shoves out the cars carrying the hot 

FIG. 94. 

charcoal from the previous charge, into a large iron box or cooler, 
placed directly opposite, where they cool out of contact with the air. 
This transfer and introduction of a new charge requires only a few 
minutes, so these retorts are practically continuous in action and the 
loss of charcoal by combustion is small. This system consumes less 
fuel, and has lower labor and repair costs for a given output than 
the small retorts. 

Movable retorts are vertical boiler-plate cylinders (Fig. 95 *) so 
arranged that the retort (A), filled with wood and with cover luted 
on, is lowered by a crane into the furnace; 
when carbonization is finished, the retort 
containing the charcoal is lifted out to cool 
unopened, while another charged with wood 
is put into its place. Connection with 
the condenser is made by a copper swing- 
pipe (B), clamped to the vent on the top of 
the retort. Each cylinder holds one cord 
of wood, and as it is packed cold, a com- 
plete filling of the space is possible. The 
labor cost for this type is less than for 
small horizontal retorts, but the wear and 
tear on the furnace and retorts from the 
frequent moving and cooling is great; the original cost of plant is 
also higher. 

Coniferous woods are often distilled in retorts into which super- 
heated or free steam is introduced, preliminary to the distillation 
proper. Thus the turpentine and much rosin are driven out at tem- 

* J. Soc. Chem. Ind., 1897, 668. 

FIG. 95. 


peratures below that at which the cellulose is decomposed. Then by 
raising the heat of the retort, true destructive distillation follows, 
yielding wood vinegar, tar, and charcoal. Many plants for distilling 
fat pine, " light wood," have been erected in this country, but not 
always with satisfactory results. Considerable wood turpentine is 
produced, the yield being 10 to 20 gallons of crude turpentine per cord 
of " light wood," but the odor and color are often not satisfactory. 

Extraction processes, depending on the use of solvents for both 
turpentine and rosin, such as carbon disulphide or carbon tetrachloride 
or turpentine itself, have been patented. The extracted chips are 
steamed to recover the solvent, but the losses are so large that the 
future of these methods is uncertain. The extracted chips are suitable 
for wood pulp, or they may be distilled for pyroligneous acid and tar. 
By treating the chipped wood with caustic soda liquor and steaming, 
the turpentine can be distilled off, while the rosin dissolves in the alka- 
line liquor. When drawn off and acidified, the rosin is precipitated. 

Pyroligneous acid or crude " wood vinegar " is a reddish brown 
liquid with strong acid reaction and empyreumatic odor due to fur- 
furol in part. It averages 5 to 10 per cent acetic acid, 1.5 to 3 per 
cent methyl alcohol, and 0.1 to 0.2 per cent acetone ; its specific gravity 
varies from 1.020 to 1.050. A small amount is used directly for mak- 
ing an impure iron acetate, sold as " pyrolignite of iron " ; but it is 
usually worked for methyl alcohol and acetic acid. 

By neutralizing pyroligneous acid directly with milk of lime and 
distilling, a raw wood spirit is collected as distillate, while calcium 
acetate solution remains in the still, which on evaporation to dryness 
yields so-called " brown acetate of lime," averaging about 67 per cent 
of calcium acetate. During the evaporation, tar separates as a scum, 
which is skimmed off. The method is not much used at present. 

By distilling pyroligneous acid in a copper still, the tar is left and 
a purified " wood vinegar " obtained, containing acetic acid, methyl 
alcohol (wood spirit), methyl acetate, acetone, acetaldehyde, etc., 
and traces of tar, empyreumatic matter, etc. This is neutralized with 
lime, precipitating many of the impurities; the clarified solution is 
then rectified in a column still, yielding wood alcohol of about 82 per 
cent. The solution in the still yields on evaporation to dryness 
" gray acetate of lime," averaging 80 per cent calcium acetate. The 
tarry matter decomposes during the drying. 

The raw wood alcohol is further purified by diluting with water 
until the oily matters (ketones, aldehydes, etc.) precipitate, as shown 


by the milky appearance. On standing several days, the oils rise to 
the top and are skimmed off; then the alcoholic solution is again 
distilled in a fractionating still, until the concentration is about 95 
per cent alcohol; the product is known as wood spirit or methyl 
alcohol. By filtering through charcoal, the color and unpleasant 
odor can be largely removed. Treatment with caustic lime and re- 
distillation yields alcohol of 99 per cent, or higher, concentration; 
but this does not remove the acetone. 

The wood spirit may be purified from acetone by treatment with 
caustic soda and iodine, producing a precipitate of iodoform with the 
acetone. Or calcium chloride is added to combine with the methyl 
alcohol and form a crystallized solid, stable at 100 C., from which 
the acetone is distilled off; then by adding hot water and heating to 
100 C., or over, the alcohol is distilled off and rectified. 

Commercial methyl alcohol is often slightly yellowish in color 
and frequently has a disagreeable odor. It is much used as a solvent 
in varnish making, for which purpose the presence of acetone is desir- 
able ; for making formaldehyde ; for mixing with ethyl alcohol to pre- 
pare " denatured ethyl alcohol " or " methylated spirit " (p. 460). 

Acetone, when recovered from wood spirit, is generally distilled 
from the calcium chloride compound with the methyl alcohol. It is, 
however, more commonly made by the dry distillation at 290 to 
300 C., of calcium acetate : 

Ca(C 2 H 3 O 2 ) 2 = CaCO 3 + CH 3 CO CH 3 . 

The product obtained by either of the above methods is crude ; sodium 
bisulphite is added, and forms a double salt with the acetone, which 
is purified by recrystallization from aqueous solution. This salt is 
decomposed by heating with sodium carbonate solution, liberating 
the acetone, which distils off pure. Or the crude acetone is neutral- 
ized with lime, settled, and the supernatant liquid diluted with water 
and rectified in a column still, yielding a pure acetone distillate and 
an oily residuum called acetone oil which finds some use for solvent 
and denaturizing purposes. 

A method for making acetone devised by Dr. E. R. Squibb * 
consists in passing acetic acid vapor through a rotating iron cylinder, 
heated to about 500-600 C., and containing pumice stone with pre- 
cipitated barium carbonate : 

= H 2 O + CH 3 CO CH 3 + CO 2 . 

* J. Am. Chem. Soc., 17, 187. 


The barium carbonate acts as a contact body, since the temperature 
is always above that at which barium acetate decomposes. The vapors 
from the still pass to a fractional condenser to remove water and acetic 
acid; the acetone condenses in a second condenser. 

Acetone is a colorless mobile liquid, having a peculiar odor and 
unpleasant taste; its specific gravity is 0.797 at 15 C. and should 
not exceed 0.802 in the commercial product. It boils at 56.3 C., and 
mixes with water in all proportions ; is an excellent solvent for many 
resins, gums, fats, nitrated cellulose, and other substances. It is used 
in the manufacture of celluloid, smokeless powders, chloroform, iodo- 
form, sulphonal, for extracting resin from crude rubber, and for 
denaturizing ethyl alcohol. 

Commercial acetic acid is prepared from gray or brown acetate of 
lime * (p. 304) by distilling with strong hydrochloric or sulphuric 
acid. In the hydrochloric acid process, copper stills heated by steam 
coils are used; free steam can also be blown into the charge. The 
acetic acid distils over, leaving calcium chloride in the still. The 
acid is a slightly colored liquid containing from 30 to 50 per cent of 
anhydrous acid, according to the strength of the hydrochloric acid 
used. It may be further purified by distilling again over a little 
potassium permanganate, and filtering through fresh charcoal. 

The sulphuric acid process is more commonly used at present; 
the still is made of cast-iron (Fig. 96 |) with a scraping device inside 
to break up the solid mass which forms in 
the still, and facilitate the escape of the acetic 
acid. The stills are heated by direct fire. A 
dust chamber should be placed between the 
still and the copper worm condenser. Owing 
to secondary reactions, some sulphuric acid 
is reduced, contaminating the product with 
sulphur dioxide, and necessitating the use of 

some excess of sulphuric acid in the still. Von Linde employs vacuum 
in the still, whereby the temperature is lowered sufficiently to permit 
steam heat to be used for the distillation, and secondary reactions 
are much reduced. 

Behrens' process consists in dissolving the calcium acetate in 
acetic acid, and then decomposing the solution with sulphuric acid, 
whereby the reaction takes place at lower temperature. 

* Brown acetate is calcined at 230 C., to destroy tarry matter before use in 
this way. 

t After Klar, Technologic der Holzverkohlung. 


The distillate contains about 75 per cent anhydrous acetic acid 
and usually a little sulphur dioxide. This acid is then rectified in a 
large copper column still, heated by a steam coil. The plates in the 
column are often porcelain or earthenware ; the condensers are usually 
copper worms. If air is excluded from the apparatus, there is little 
attack on the copper by the acid ; but this necessitates the immediate 
refilling of the still after each charge has been worked off. If the 
operation is to be discontinued, the still and condensers should be 
thoroughly washed out with water. When starting each distillation, 
the heating should be slow, to allow the sulphur dioxide to pass off 
before the acetic acid begins to distil. 

According to the amount of cooling water admitted to the frac- 
tional condenser, a clear colorless liquid containing from 80 to 99 
per cent of anhydrous acid can be obtained; it contains traces of 
empyreumatic matter, which can be removed by gentle heating with 
potassium permanganate in an earthenware vessel, and redistilling 
in a copper still, with an earthenware, or pure silver, worm condenser. 
The residues from the several distillations are collected together and 
redistilled to recover as much as possible of the acetic acid in them. 
The final tarry residues are burned. 

By distilling the pyroligneous acid, without neutralization, in a 
copper still, most of the methyl alcohol passes over before the acetic 
acid ; by collecting the distillate until its specific gravity is about 
1.000, a crude " wood spirit " is separated. If the acetic acid vapors 
following are passed into a solution of soda, a solution of sodium 
acetate is obtained. This is evaporated until only a fused mass of 
sodium acetate remains, which is heated to nearly 300 C., at which 
point the sodium acetate is stable but the impurities are decomposed. 
The fused salt is dissolved in water, the solution filtered and evaporated 
to crystallize. The process may be repeated for further purification. 

If the distillation of the pyroligneous acid be continued after the 
methyl alcohol has passed over, the distillate, collected between 100 
and 120 C., is called " wood vinegar " ; it is dilute and still retains 
some empyreumatic matter, but is somewhat employed technically. 
It is generally neutralized with lime or soda to yield acetates. 

Glacial acetic acid is the nearly anhydrous, 99 to 100 per cent 
acid, which crystallizes if cooled to 16.5 C. It may be made from 
fused sodium acetate, by distilling with strong sulphuric acid at 120 C. 
The residue from the still is sodium sulphate and may be used to 
decompose calcium acetate in solution, to prepare more sodium 


Common acetic acid is found in commerce as a slightly colored 
liquid of various strengths : ordinary No. 8 has a specific gravity about 
1.040 (8 Tw.) and contains approximately 30 per cent anhydrous 
acid ; it is used in preparing acetates, in making white lead, in textile 
work, and in pharmacy. Stronger acid, containing 50 per cent or 
more of anhydrous acid, is used in preparing coal-tar colors and calico 
printers' pastes, for preparing organic acetates, and for solvent pur- 
poses. Pure acetic acid from wood distillate may be used for vinegar, 
but lacks the characteristic salts and flavoring substances present in 
true fermentation vinegar (p. 463). 

Acetates. Aluminum acetate in the pure state is not known, but 
a solution of it in acetic acid, called " red liquor," is largely used in 
dyeing and in calico printing. It is made by dissolving aluminum 
hydroxide in acetic acid, or by decomposing lead or calcium acetates 
with aluminum sulphate or alum : 

A1 2 (S0 4 ) 3 + 3 Pb(C 2 H 3 O 2 )2 = 2 A1(C 2 H 3 O 2 )3 + 3 PbSO 4 . 

Calcium acetate yields the best red liquor; that made from lead 
acetate is not entirely free from lead, which dulls the shade of deli- 
cate colors ; when made from alum, it contains sulphate of the alkali 
metal, and decomposes more readily than when made from aluminum 
sulphate. Several basic aluminum acetates are made by adding 
sodium carbonate to the normal acetate solution. These deposit 
alumina on the fibre very readily. 

Chromium acetate finds some use as a mordant in calico printing. 
It is usually made by dissolving chromium hydroxide in acetic acid, 
or by decomposing a solution of chromium sulphate or chrome alum 
with lead or calcium acetate. The solution is violet, but becomes 
green if heated. It may be evaporated to dryness without rendering 
the salt insoluble. Alkalies and alkaline carbonates yield no precipi- 
tate in the cold solution, but when heated, a precipitate of chromium 
hydroxide forms. 

Basic acetates are prepared by adding lead or calcium acetate to 
basic chromium sulphate solution. Sulphate-acetates are also made 
and used as mordants. 

Calcium acetate has been mentioned as brown or gray acetate of 
lime (p. 304). The pure salt, occasionally used as a mordant, is 
made by neutralizing acetic acid with the theoretical quantity of 
lime. Litmus does not show the point of neutrality. The crystal- 
lized salt, Ca(C 2 H 3 O 2 ) 2 H 2 O, is very soluble in water. 


Cupric acetate, Cu(C 2 H 3 O 2 )2 H 2 O, is best made by adding lead 
acetate to copper sulphate solution : 

CuS0 4 + Pb(C 2 H 3 O 2 )2 = Cu(C 2 H 3 O 2 ) 2 + PbSO 4 . 

It may be made by dissolving verdigris, or copper carbonate or oxide, 
in acetic acid. For basic acetates see p. 237. 

Ferrous acetate, Fe(C 2 H 3 O 2 ) 2 4 H 2 O, may be prepared from cop- 
peras and lead or calcium acetate; or by dissolving scrap iron in 
acetic acid. It is quickly oxidized in the air to basic ferric acetate. 
" Pyrolignite of iron," black liquor, or iron liquor, is made by dissolv- 
ing scrap iron in pyroligneous acid. It is sold as a dirty olive-brown 
or black liquid, having a density of about 25 Tw., and consists mainly 
of ferrous acetate, with some ferric acetate and tarry matter. It 
is used as a mordant in dyeing black silks and cottons, and in calico 

Ferric acetate, Fe(C 2 H 3 O 2 ) 3 , made by adding lead acetate to fer- 
ric sulphate, is stable in cold solution. It forms basic salts when 
treated with caustic soda. It was formerly used in black silk dyeing. 

Sodium acetate, NaC 2 H 3 O 2 3 H 2 O, forms needle-like crystals 
which melt in their crystal water when heated ; when anhydrous, it 
fuses without decomposition. It is chiefly used for making pure 
concentrated acetic acid, in making certain diazo bodies, and as a 
developer for the azo-dyes, in which the color is made on the fibre. 

Lead acetate, Pb(C 2 H3O 2 ) 2 3 H 2 O, " sugar of lead," is made by 
dissolving feathered lead by causing acetic acid to trickle over it in the 
presence of a current of air. Or litharge is dissolved directly in acetic 
acid. If wood vinegar is used, the product is " brown sugar of lead." 
With an excess of litharge, basic acetates are formed. The normal salt 
is very soluble in water, and is used for making other mordants and for 
chrome yellows. The salts are poisonous, and are affected by the car- 
bon dioxide and hydrogen sulphide in the air. 

Wood-tar varies somewhat in character with the kind of wood 
carbonized. It is washed with hot water, or treated with milk of 
lime, to remove acetic acid, and then washed with very dilute sul- 
phuric acid. Excess of water is evaporated by warming in steam- 
jacketed vessels. The tar is then distilled in iron stills, provided 
with stirring apparatus, the temperature being raised very slowly. 
The distillate collected below 150 C. is called " light oil," and is 
chiefly used as a substitute for oil of turpentine in varnish and paints. 
Between 150 and 250 C. the " heavy oil " is collected, containing 


creosote, toluene, and paraffine bodies. By stopping the distillation 
at 250 C., a thick, brownish liquid is obtained, which is used in 
making axle grease, shoemakers' wax, for lampblack, and for coating 
the interior of casks and barrels to render them impervious to liquids. 

The creosote oil is washed with caustic soda, and boiled in the air 
to oxidize various substances which it contains. The alkaline solu- 
tion is then acidified with sulphuric acid, to precipitate the creosote, 
which is treated with alkali and acid as before. It is then distilled 
again, and the distillate, collected between 200 and 220 C., is the 
commercial wood-tar creosote. It has a strong, smoky odor, is a 
good antiseptic, and ig not poisonous. 

Stockholm tar and pine tar are obtained by a crude distillation of 
pitch-pine or other coniferous wood, in heaps, covered with turf. 
These are of different composition from retort tar, and are mainly 
used for tarred ropes, with oakum for ship calking, and for preserving 


Das Holz und seine Distillations-Producte. Dr. G. Thenius, Leipzig, 1880. 

Die Meiler und Retorten Verkohlung. Dr. G. Thenius. 

Das Chemische Technologie der Brennstoffe. F. Fischer, Braunschweig, 

Die Verwerthung des Holzes auf chemischen Wege. J. Bersch, Leipzig, 

Destructive Distillation. E. J. Mills, London, 1892. (Gurney and 

Handbuch der Organischen Chemie. Victor Meyer and Paul Jacobson. 

Vol. I. Articles " Essigsaure " and " Methylalkohol." Leipzig, 


Technologie der Holzverkohlung. M. Klar, Berlin, 1910. (Springer.) 
Handbuch der Organischen Chemie. F. Beilstein. Vol. 1, 3d ed. Puri- 
fication of Wood Spirit. Leipzig, 1894. (L. Voss.) 
Jahres-Bericht tiber die Leistungen der technischen Chemie : 

1892. 1893, 14. (Distillation of Wood.) 
Journal of the Society of Chemical Industry : 

1892, 395 and 872. 1897, 667, 722. M. Klar. (Modern Distillation 

of Wood.) 


Bones are usually extracted with benzine or with carbon disul- 
phide, and the fatty matter used for soap stock. They still contain 
nitrogenous organic substances, and are distilled in iron or clay 
retorts, similar to those used in coal-gas making (p. 315), yielding 
volatile products, consisting of gases, ammonium salts, and bone oil ; 
these pass through condensers, where the water and bone oil con- 
dense; the gases pass into a receiver containing sulphuric acid, 
which takes up the ammonia and its volatile compounds ; the in- 
flammable gases are burned under the retort. 

The bone oil (" Dippel's oil ") and aqueous liquor collected under 
the condensers are separated by gravity. The liquor contains am- 
monium carbonate, cyanide, sulphocyanide, and sulphide, and is 
treated in the same way as gas liquor (p. 151) for the recovery of the 
ammonia. The crude bone oil is a 'dark-colored, foul-smelling liquid, 
lighter than water. It is redistilled and divided into numerous 
fractions. At high temperatures it also yields ammonium carbonate 
and cyanide ; the thick tar remaining in the still is the basis of com- 
mercial Brunswick black. 

The constituents of bone oil are exceedingly numerous, but the 
more important are pyrrol, C 4 H 4 NH ; pyridine, CsHsN; picoline, 
C.5H 4 (CH3)N; lutidine (dimethylpyridine) ; collidine, C 5 H 2 (CH 3 )3N; 
and quinoline, CeH^ CaHaN. These have but little technical use, but 
are employed in Europe for denaturating alcohol, and in the prepara- 
tion of certain antiseptics. They are closely related to some of the 
alkaloids, but are not as yet used to prepare them. 

The residue from the bone distillation is the bone-black or bone- 
char of commerce. It forms about 65 per cent of the original weight 
of the bones and consists largely of calcium phosphate and carbonate, 
impregnated with free carbon. While still hot, it is drawn from the 
retort into closed vessels and cooled out of contact with the air. It is 
largely used in decolorizing sugar solutions, glucose, glycerine, oils, 
paraffine, vaseline, etc., and in case-hardening iron. It loses its effec- 
tiveness after a time, and is then " revivified " (p. 418). When it 
becomes too finely powdered for successful filtration, it is used as a 
fertilizer (p. 165). . 



Illuminating gas may be made by enriching water gas with oil 
gas, or by the destructive distillation of coal, wood, or petroleum. 
Goal gas, such as is generally used at the present time, was first 
employed for house illuminating by William Murdock, in London, in 
1792. It was introduced for street lighting in London in 1812, and 
in Paris in 1815. In this country, the so-called water gas, enriched 
with naphtha, has largely replaced coal gas in many of the large 
cities. This has greater illuminating power, requires a smaller plant 
and less labor, and ensures greater economy of working. 

Water gas (p. 39) is produced by the action of steam on incandes- 
cent carbon, according to the reactions : 

C + 2 H 2 = 2 H 2 + CO 2 . 
= 2 CO. 

It is composed chiefly of hydrogen and carbon monoxide, is non- 
luminous, and has a high heat value. 

Luminosity depends on the presence of hydrocarbons, such as 
ethane, C 2 H 6 , ethylene (ethene), C 2 H 4 , acetylene, C 2 H 2 , and benzene, 
CeHe, and their homologues, the most important of these " illu- 
minants " being ethylene and benzene. In order to render the water 
gas luminous, it is carburetted with gases derived from oil, which are 
rich in illuminants. 

Illuminating water gas . is now made by two general methods : 
(a) the carburetted gas is made in one operation ; (6) non-luminous 
gas is prepared, and then carburetted by a second process. The first 
method is most successfully carried out by the Lowe process. The 
generator (Fig. 97) is filled with anthracite coal or coke, which is 
brought to incandescence by a blast of air. The gases from the 
generator, at this time consisting mainly of carbon monoxide and 
nitrogen, enter at the top of the carburettor, a circular chamber 
lined with firebrick, and containing a " checker-work," of the same 
material ; while passing down through this, the producer gas (p. 41) 
is partly burned by an air blast which enters the apparatus near the 
top, and the checker-work is heated white hot. The gases pass on 
to the " superheater," a taller chamber, also filled with checker-work. 
At the bottom of this an air blast is introduced to complete the 
burning of the producer gas and to raise the temperature of the 




checker-work to a very bright red heat. From the top of the super- 
heater, the waste gases escape into a hood leading into the open air. 
When both the carburettor and superheater have reached the desired 
temperature, the air blasts are cut off, and steam is introduced into 

FIG. 97. 

the generator, where it is' decomposed by the incandescent fuel, 
according to the reactions. The water gas thus formed passes into 
the carburettor, while a small stream of oil is being introduced through 
a pipe at the top. The oil is decomposed by contact with the hot 
checker-work, forming illuminating gases which mix with the water gas, 
and passing into the superheater, are completely fixed as non-condens- 
able gases. 

It is customary to run the air blast for some eight minutes, when 
the fuel reaches a temperature of about 1100 C. The steam, super- 
heated before entering the generator, is run about six minutes, until 
the temperature of the generator and carburettor has fallen below 
the point at which decomposition occurs. In order to economize 
heat, the hot carburetted gas is passed through a pipe surrounded 
by a jacket, within which the oil is circulating, thus heating it be- 


fore it enters the carburettor. The lower end of the pipe leading 
from the superheater is closed by a water seal, to prevent any back- 
ward rush of the gas during the operation of the air blast. It is cus- 
tomary to lead the gas from the superheater into a storage holder, 
from which it is drawn through the purifying apparatus. 

In this process, the blowing of air and of steam are intermittent, 
but the actual formation of gas is accomplished in one operation. 

The second method of preparing illuminating water gas is the 
Wilkinson process. Water gas is made by blowing steam into the 
hot coal in the generator, and is stored in the holder. A measured 
quantity of gas is then introduced into the carburettor, a closed iron 
box, containing slightly inclined plates, over which the exact amount 
of oil necessary to carburet the gas, is flowing in very thin layers. 
The carburettor is also provided with a steam jacket and coils to 
keep the temperature high enough to vaporize the oil. These vapors 
mix with the gas and pass at once into the fixing apparatus, which 
is a long, narrow, fire-clay retort, kept at a white heat by external 
fire. Here the oil vapors are " cracked " into hydrocarbons, which 
are non-condensable gases, and being mixed with water gas, render 
it luminous when burned. The mixed gases then go directly to the 
scrubbers and purifiers. For 1000 cubic feet of gas, about 50 pounds 
of anthracite and 4.2 to 5 gallons of naphtha are consumed. 

The impurities in the water gas are essentially the same as those 
in coal gas, and the method of washing and purifying are the same. 

The illuminating value of coal gas is frequently raised by mixing 
it with carburetted water gas. Owing to its high percentage of car- 
bon monoxide, water gas is exceedingly poisonous when inhaled, and 
much care is necessary to prevent leakage into inhabited rooms (see 
table, p. 325). 

Coal gas, prepared by the destructive distillation of bituminous 
coal, is generally made by the smaller gas companies in this country. 
In Europe scarcely any water gas is made for illuminating purposes. 
The composition and yield of coal gas depend upon the kind of coal 
used and the manner of distillation. A " fat " coal, moderately low 
in sulphur, and caking on distillation to a good coke (e.g. the Penn- 
sylvania gas coals), is most desirable for illuminating gas. The 
temperature of the retort is a very important factor in the character 
of the distillation products. When it is low, the quantity of gas 
formed is small, but it contains a large percentage of ilium inants, 
and hence is of a high candle power. When the temperature in the 


retort is high the effects are as follows : (a) the yield of gas is 
much increased, but the percentage of methane, ethane, and hydrogen 
is much greater, and since these have very little illuminating value, 
the gas is of low candle power; (b) the yield of tar is increased; 
(c) the vapors of the heavy hydrocarbons which constitute some of the 
tar are decomposed on coming in contact with the hot retort, form- 
ing gases of lower carbon content, and depositing free carbon on its 
walls. This " gas carbon " * adheres very firmly and if allowed to 
become thick causes much loss of heat. It is especially liable to 
deposit if there is undue pressure in the retort, which may be the 
case if the exhausters are not working properly ; (d) there is a larger 
yield of organic bodies having ring nuclei in their composition, such 
as naphthalene, phenols, anthracene, etc. These not only cause loss, 
but also cause clogging in the service pipes and burners. 

The products of the distillation are gas, ammoniacal liquor, tar, 
and coke. When coal is distilled for coke (p. 35), the ammoniacal 
liquor and tar are sometimes saved by the use of by-product ovens, 
but the gases are burned for fuel or go to waste. When distilled for 
illuminating gas, the process is carried on with a view to the best 
yield of high quality gas, but the ammoniacal liquor, tar, and coke 
are valuable by-products. The coke is too soft for metallurgical pur- 
poses, and is chiefly used to heat the retorts or sold for domestic fuel. 

A diagram of a complete plant for coal gas making is shown in 
Fig. 98. The retorts (A) are Q-shaped, fire-clay vessels, about 8 
feet long, 18 inches wide, and 15 inches high; they are set six or 
eight together in a furnace, the whole constituting what is called a 
" bench." Each retort has a cast-iron mouthpiece projecting out of 
the furnace, and carrying the door, closed by a screw clamp. Re- 
torts may be " single," i.e. closed at one end and having but one 
door for charging and discharging ; or they are " through " retorts, 
about 18 feet long, having a door at each end, so that they may be 
charged or emptied from either side of the furnace. A modified 
form of the latter is the " inclined " retort, set at an incline of about 
32, the coal being run in at the upper end, and the coke discharged 
by gravity, by opening the door at the lower end. Vertical retorts 
are also in use. Each bench is heated to 1000 or 1200 C. by a coke 
fire on a grate below the retorts, or, in more modern plants, by gen- 
erator gas. A number of benches are built together, and constitute a 
" stack." 

* Gas carbon is used for electric light carbons, battery plates, and other electri- 
cal appliances. It is denser and purer than most other forms of carbon. 



FIG. 98. 

From the front of each retort a vertical cast-iron pipe (B) about 
6 inches in diameter, and called the " stand-pipe," ascends to the 
top of the bench, where it joins the " bridge-" and " dip-pipes," 
which conduct the volatile products from the retort to the hydraulic 
main (C). This is a long covered trough, extending the entire length 
of the stack, and receiving the gas .and distillate from each retort. 
In it the greater part of the tar and oily products condense and collect 
under the water which is kept in the main to act as a seal to the ends 
of the dip-pipes, to prevent the gas from passing back into the retort 
when the latter is opened. Ammonium salts, such as sulphate, sul- 
phide, and carbonate, are washed from the gas as it bubbles through 
the water, and are afterwards recovered (p. 151). The ends of the 
dip-pipes must not extend more than 2 inches into the water ; other- 
wise, there is pressure in the retort and consequent loss from leakage 
and from deposition of carbon in the retort and stand-pipe. If the 
gas is allowed to cool in contact with the tar, the latter absorbs some 
of the illuminants, thus reducing the candle power (p. 323). If the 
stand-pipes are too hot, the volatile constituents of the tar are driven 
out, and a very thick mass deposits, causing clogging. 

From the hydraulic main a pipe (not shown) leads to the con- 
denser (D), which consists of a series of vertical cast-iron pipes, con- 
nected by bends at the top, and opening at the bottom into an iron box. 


This box is divided by transverse partitions which do not extend to 
the bottom, merely dipping into the ammoniacal liquor and tar con- 
tained in it. The liquor forms a seal, thus forcing the gas to pass 
through the pipes, while the condensed products flow along the bottom 
of the box to the tar well. These condensers are simply air cooled, 
but certain forms are constructed with water coolers. In those most 
frequently in use in this country the pipes are laid at a slight incline 
to the horizontal. 

The annular condenser consists of a series of vertical pipes con- 
nected by diagonal pipes leading from the bottom of one to the top 
of the next. Through each of these vertical pipes a smaller tube 
passes parallel to the length of the pipe and opening to the air at 
both ends, thus forming an annular space in each pipe, through 
which the gas passes downward, and then through the diagonal pipe 
to the top of the next. In this way a very large air-cooled surface 
is obtained. At the bottom of each cooling pipe a small pipe car- 
ries away the condensed tar and liquor. 

The tubular condenser consists of a rectangular box about 2 feet 
wide and 20 feet high, divided into narrow sections by partitions 
extending alternately to within a few inches of the top and of the 
bottom. Through each section, a number of narrow horizontal tubes, 
open to the air at each end, extend from one side of the box to the 
other. In this way the gas passing through the sections is exposed 
to a very large cooling surface. 

Water condensers consist essentially of pipes surrounded by flow- 
ing water. Through these the gas is made to pass in a direction 
opposite to that in which the Water flows. By regulating the supply 
of water, the temperature is easily controlled. 

The object of the condenser is to cool the gas slowly to the tem- 
perature of the atmosphere, provided this in not under 50 C. Cool- 
ing below this causes condensation of some of the illuminants, with 
corresponding loss. If the cooling is very rapid, the tarry matter 
separates quickly, and drags some of the lighter hydrocarbons down 
with it. 

The exhauster (E, Fig. 98) draws the gas from the retort, through 
the hydraulic main and condenser, and acts as a pump forcing it 
through the remaining parts of the plant. By drawing the gas out 
of the retort quickly there is less decomposition of the gas itself, 
and hence less carbon is deposited in the retort; a larger yield 
results and less fuel is necessary, while the retort lasts longer. 

Another form of exhauster is a direct-acting pump, which draws 


the gas from the retort and condenser, and forces it to the 

Root's rotary exhauster is frequently employed, as is also Beal's 
(Fig. 99). This consists of an outer circular casing having inlet and 
outlet pipes, and an inner revolving drum (B), turning on an eccentric 
axis in such a way that the drum just touches the lowest point of the 
inner surface of the casing. Through slots cut in the drum, two 
blades or diaphragms (D) slide freely over one another, to form a 
double diaphragm, variable in width, according to the relative posi- 
tion of the blades to each other. In the outer end of each blade is a 

FIG. 99. FIG. 100. 

pin, which travels in a circular groove sunk in the ends of the casing. 
Thus as the drum revolves about its axis, the pins, travelling in the 
fixed groove, draw the blades in and out, across the axis of the drum. 
The outer ends of the blades are thus always kept in contact with the 
walls of the casing. The exhauster is driven by an engine, and the 
rotary blades and drum catch the gas which enters through the inlet, 
and force it out through the other pipe. 

The steam jet exhauster (Fig. 100 *) is effective, but heats the 
gas, which is afterwards cooled in the washer. A jet of steam is 
blown through conical openings into a wide pipe, drawing the gas 
along with it into the cones. 

The tar extractor (F, Fig. 98) is a short tower filled with numer- 
ous horizontal perforated plates. The friction of the gas in passing 
through the small holes in these plates removes the last traces of tar and 
prevents clogging in the scrubber. In Europe the apparatus of Pelouze 
and Audouin is much employed. This is a bell made up of three layers 
of wire netting, or of perforated plates, which is suspended in a water 
seal. The gas enters under the bell and passes through the meshes 

* After Ost, Lehrbuch d. tech. Chemie. 


or perforations of the bell walls, to which the tar particles attach 
themselves and finally drop to the bottom and run off by a special 

The scrubber and washer are intended to remove the ammonia 
and part of the carbon dioxide and hydrogen sulphide. In the 
former the gas is brought into contact with thin films or layers of 
ammoniacal liquor from the hydraulic main or condensers, which 
trickles over coke, twigs, wooden slats, or pebbles, in a tower. This 
liquor absorbs some of the carbon dioxide and hydrogen sulphide, 
which combine with the ammonia. In the washer the gas is brought 
in contact with pure water, trickling over twigs, coke, etc., and 
which removes the ammonia from the gas. 

Tower scrubbers are tall cast-iron vessels built in segments, each 
of which has a " grid " or grating, upon which the filling material 
is supported. Two towers are always used in conjunction, the first 
fed with ammoniacal liquor and the second with water. The amount 
of liquor and water is carefully regulated, and the gas entering at 
the bottom of the first tower passes up and then to the bottom of 
the second, and is thus first brought into contact with the strongest 
liquor and finally with pure water. These tower scrubbers are now 
only used in old plants ; in all modern establishments they have been 
replaced by scrubber-washer machines. 

The Standard scrubber-washer machine (G, Fig. 98) is a Q-shaped 
iron vessel divided into a series of narrow chambers by transverse 
partitions. In the upper half of the apparatus is a revolving shaft 
carrying a number of thin wooden grids, bolted together in parallel 
segments, with blocks making a space of about one-eighth of an inch 
between each pair of grids. A group of these slats revolve in each 
chamber. Water at about 60 F. is admitted to the last chamber of 
the series, at the rate of about one gallon for each 1000 cubic feet of 
gas, and, automatically regulated, flows from chamber to chamber in 
a direction opposite to that in which the gas is passed. Thus the 
fresh water comes in contact with the most nearly purified gas. The 
level of the water is lower in each succeeding chamber, until in the 
first chamber, where the gas enters the apparatus, the strong ammo- 
niacal liquor is only a few inches deep. 

By the revolution of the shaft, the grids are submerged in the 
liquor, and freshly wetted surfaces are brought into the urjper part 
of the apparatus. By a suitable arrangement of baffle plates, the 
gas is made to enter each chamber at the centre, and find its way to 
the circumference by passing through the narrow spaces between the 


grids. The water, forming a thin film on them, absorbs the ammonia, 
carbon dioxide, etc. ; and as the shaft revolves from 12 to 15 times 
per minute, the solution formed is at once mixed with the liquor in 
the bottom of the chamber. The machine works effectively, and in 
this country is rapidly replacing the tower scrubbers. 

From the scrubbers the gas passes to the purifiers (H, Fig. 98), 
whose chief purpose is to remove sulphur compounds. They are 
shallow rectangular iron boxes, each having a false bottom, upon 
which the purifying material rests. The gas enters under this grating 
and leaves by a pipe opening just under the cover, which rests in a 
hydraulic seal, and is lifted by a travelling crane. Usually four puri- 
fiers are placed in a series, one of which is emptied and recharged at 
a time, without interrupting the purification process. The foul gas 
enters the most nearly exhausted purifier, and, passing through the 
others, leaves the apparatus through that most recently charged, con- 
nection being made between the purifiers by means of a complicated 
piece of apparatus (L and 0, Fig. 98) called the centre seal. 

The purifying materials may be slaked lime or hydrated ferric 
oxide. Lime is the oldest material used and is also the best, since 
it removes both the carbon dioxide and carbon disulphide. But it is 
expensive, and the spent lime, having a most offensive odor and con- 
siderable bulk, is difficult to dispose of. The lime should be thor- 
oughly slaked several days before use, and should contain as much 
water as it will hold without becoming pasty or liquid. It is placed 
in the purifiers in layers about six inches deep. The reactions occur- 
ring with lime are : 

1) Ca(OH) 2 + 2 H 2 S = Ca(SH) 2 + 2 H 2 O. 

2) Ca(OH) 2 + H 2 S = CaS + 2 H 2 O. 

3) CaS + CSa = CaCS 3 (calcium thiocarbonate). 

4) Ca(OH) 2 + CO 2 = CaCO 3 + H 2 O. 

Since carbon dioxide will decompose calcium sulphide, sul- 
phydrate, or thiocarbonate, if gas containing it is passed through a 
foul purifier, the following is liable to take place : - 

CaS + CO 2 + H 2 O = CaCO 3 + H 2 S. 
. CaCSs + CO 2 + H 2 O = CaCO 3 + 82 + H 2 S. 

The volatile sulphides thus liberated must be removed in a second 
purifier, into which no carbon dioxide enters. Carbon dioxide has a 
deleterious effect on the illuminating power of the gas. 


When iron oxide is used, only the hydrogen sulphide is removed 
from the gas : 

1) Fe 2 O 3 3 H 2 O -t 3 H 2 S = 2 FeS + S + 6 H 2 O. 

2) Fe 2 O 3 3 H 2 O + 3 H 2 S = Fe 2 S 3 + 6 H 2 O. 

The oxide is a natural bog iron ore, Fe 2 O 3 3 H 2 O. When fresh, 
it contains about 50 per cent water and a large amount of vegetable 
matter, but before use it is dried until about one-half of the moisture 
is expelled, and is then mixed with an equal bulk of sawdust to ren- 
der it more porous. When it becomes inactive through absorption of 
sulphur, it is " revivified " by removing it from the purifier and 
spreading it in a layer a foot or more in depth, where the air can 
act upon it. Considerable heat is evolved by the action of the oxy- 
gen of the air on the iron sulphides : 

2 FeS + 3 O = Fe 2 O 3 + 2 S. 
Fe 2 S 3 + 3 O = Fe 2 O 3 + 3 S. 

Thus free sulphur is deposited in the oxide. The ore may be revivi- 
fied repeatedly until the free sulphur accumulates in it to the amount 
of 50 or 55 per cent, when the proper action in the purifiers is hin- 
dered, and fresh oxide must be used. If some air is admitted along 
with the gas, the iron oxide is revivified in the purifiers, and need 
not be removed so often ; but this dilutes the gas slightly with nitro- 
gen. One ton of good iron oxide will purify ten to twelve million cubic 
feet of gas. Sometimes lime is used before the iron oxide, in order 
to remove carbon dioxide. Any sulphur compounds of the lime 
which may be formed are decomposed by the carbon dioxide in the 
foul gas (see above). Considerable carbon dioxide is present in 
unpurified water gas, and is generally thus removed before the gas 
enters the iron oxide purifiers. 

The purified gas passes through the station meter (I, Fig. 98) and 
then to the holder (J), from which it is delivered to the street mains. 

The Feld process of gas purification has found some favor in 
Europe. The gas, which has not been freed of ammonia, is washed in 
a scrubber with a dilute solution of ferrous sulphate ; both ammonia 
and hydrogen sulphide are absorbed : - 

FeSO 4 4- 2 NH 3 + H 2 S = FeS + (NH 4 ) 2 SO 4 . 

The exhausted wash-liquor is regenerated by blowing sulphur dioxide 
gas into it, forming soluble ferrous thiosulphate and precipitating sul- 
phur : - 2 FeS + 3 SO 2 = 2 FeSsOg + S. 


This liquid, with the sulphur in suspension, is used in the scrubber 
for more crude gas : 

+ 2 NH 3 + H 2 S = FeS + 

This liquor is in turn regenerated by blowing a mixture of air and 
sulphur dioxide into it, when the thiosulphate is oxidized to sul- 
phate : 

2 FeS + 2(NH4) 2 S2O 3 + 3 SOa + 2 O 2 = 2'FeSO 4 + 2(NH 4 ) 2 SO 4 +5 S. 

This is repeated until the ammonium sulphate reaches 30 to 40 per 
cent, when the sulphur is filtered off ; the iron precipitated as ferrous 
sulphide is also filtered and the sludge returned to the process. The 
ammonium sulphate is recovered by evaporation. 

Cyanides (p. 289) are recovered from the impure gas in some of 
the large works. When this is done by the use of iron salts,* with- 
out previous removal of the ammonia, a special arrangement of the 
apparatus is desirable. The tar extractor is put behind the air- 
cooled condenser; next is a standard scrubber, charged with heavy 
tar oils in the forward compartments to remove naphthalene, and the 
iron solution is in the later compartments to remove the cyanogen; 
following this is the water-cooled condenser, and then the ammonia 
scrubber. Thus the gas enters the ammonia scrubber nearly cold. 
This removal of cyanogen from the gas renders the activity of the 
iron oxide in the purifiers of greater duration, for only the hydrogen 
sulphide is to be removed, no sulphocyanide nor Prussian blue is 
formed, and the ultimate amount of sulphur in the mass readily reaches 
50 per cent, when the material is suitable for making sulphuric acid. 

The ammoniacal liquors from the hydraulic main, condensers, 
and scrubbers are mixed, forming " gas liquor " of approximately 
" 10 ounce " strength, i.e. the ammonia gas which can be liberated 
from one gallon of the liquor will neutralize 10 ounces by weight of real 
sulphuric acid. It is used for the production of ammonia (p. 151). 
The tar from the tar well is shipped to the tar distiller (p. 327). 

The usual impurities found in gas are ammonia, hydrogen sul- 
phide, and carbon dioxide. Ammonia is detected by holding a strip 
of wet turmeric or litmus paper in a stream of the gas ; the former 
becomes brown or red, and the latter blue. For hydrogen sul- 
phide, paper wet in lead acetate or silver nitrate is used. Carbon 
dioxide is detected by shaking a small bottle of the gas, freed from 
hydrogen sulphide, with lime or baryta water. 

* Journal fur Gasbeleuchtung, 1899, 470. 


The yield from one ton of good gas coal is approximately : 

.10,000 cu. ft. 16 candle power gas. 
1400 pounds coke. 
120 pounds tar. 
20 gallons ammoniacal liquor (10 to 12 oz.). 

The illuminating power of gas is expressed in "candles," by 
which is meant the ratio of its illuminating power to that of a " stand- 
ard candle," as measured by a photometer. The English standard 
is the light of a sperm candle, weighing one-sixth of a pound, when 
burning 120 grains per hour. But this is subject to variation, and 
much ingenuity has been expended in devising a better standard. A 
burner designed for use with a mixture of air and pentane, C5Hi2, has 
found some favor in Europe. In Germany the light of a lamp burning 
amyl acetate, with a specified height of flame and size of wick, is the 
official standard. In this country standard candles are used. When 
testing gas, it is customary to burn it at the rate of 5 cu. ft. per hour, 
in a burner of the argand type. Ordinary coal gas is about 16 candle 
power, but it is sometimes " enriched " by putting into the retort, 
along with each charge of coal, an iron cylinder containing petroleum 
oil. This is closed with a cork, which burns out, and the escaping oil 
is decomposed, the vapors mixing with the coal gas, increasing its illu- 
minating power. 

Another method of enriching coal gas is the addition of benzol 
vapors, the gas having been previously scrubbed with heavy oil to 
remove naphthalene, phenols, and other constituents. 

A modern improvement in gas lighting is the introduction of in- 
candescent burners, in which the non-luminous flame of a Bunsen 
burner is made to heat a mantle or gauze composed of the oxides of 
various rare earths, especially thorium and cerium, which possess in 
high degree the property of selectively radiating light at relatively 
low temperatures. The mantle heated to incandescence glows with 
a powerful light, while very little heat is given out. These burners 
consume about three and one-half feet of gas per hour, and their 
efficiency is four times that of an ordinary argand burner. They are 
advantageous to use with a gas of low illuminating power, provided it 
has considerable heating value. 

Oil gas is now largely made by " cracking " certain petroleum, tar, 
or shale oils in retorts. In Pintsch's process the retort is divided 
by a partition into an upper and lower chamber ; the oil is cracked 
in the upper compartment, and the vapors pass into the lower one, 


which is heated to about 1000 C., where they are " fixed " and form 
permanent gases. In Peebles' process the retorts are not so hot, 
and the oil is partly cracked and partly distilled ; the heavier fractions 
condense and return to the retort, and only very volatile hydrocarbons 
leave the apparatus. Purified oil gas has a high candle power, usually 
over 50, and is burned in special forms of burner, otherwise it is liable 
to smoke or deposit soot ; it is rich in benzene and olefine hydrocarbons, 
and may be burned alone or used to enrich other gases. Burning 
with pure oxygen improves its combustion and illuminating power 
greatly. Pintsch gas is much used for lighting railroad cars; it is 
compressed into cylinders for carriage, but the pressure must be low 
or great loss of illuminating power occurs owing to the condensation 
of the heavy illuminants. 

Blau gas * is made by " cracking " petroleum in a steel retort, at 
low red heat (500-600 C.) ; both liquid and gaseous products re- 
sult, which are cooled to condense the higher boiling constituents. The 
gas then passes a purifier (iron oxide and lime) to remove hydrogen 
sulphide and carbon dioxide ; then it is compressed in a four-stage 
compressor to 100 atmospheres, whereby part of it liquefies, and under 
the heavy pressure, the liquid hydrocarbons dissolve much of the fixed 
gases and hold them in solution. The liquid is then charged into steel 
cylinders for use. The residual unabsorbed gas separated from the 
liquid is used in gas engines driving the compressors. 

Blau gas is nearly free from carbon monoxide and consists essen- 
tially of methane and hydrogen dissolved in saturated and unsaturated 
hydrocarbons, which impart high calorific and illuminating value (50 
c. p.). It is used in isolated buildings, factories, yachts,, and railway 
cars ; and with oxygen for autogenous welding and metal cutting. 

Acetylene is made from calcium carbide (p. 266) by treating with 

CaC 2 + 2 H 2 O = Ca(OH) 2 + C 2 H 2 . 

One ton of 80 per cent carbide yields about 9000 cu. ft. of acetylene 
gas. The crude gas is usually contaminated with hydrogen sulphide, 
phosphine, and ammonia ; it is purified by passing over bleaching pow- 
der, chromic acid, or cuprous chloride. When burned under pressure, 
in a special form of burner, it yields a very brilliant light. It is not 
used to enrich coal- or water-gas, since its candle power is much lowered 
by mixing with other gases. With air in greatly varying proportions 
the gas forms explosive mixtures. 

Heavily compressed acetylene gas is liable to explode; hence 

* Jour. Soc. Chem. Ind., 1908, 550. Met. and Chem. Eng., 1914, 153. 



storage in the compressed state in gas tanks is dangerous. But enor- 
mous quantities of the gas dissolve in acetone under pressure, and the 
solution is not explosive ; on releasing the pressure, the gas is evolved 
rapidly and steadily. By filling the cylinders completely full of some 
porous material, as asbestos, or charcoal embedded in cement, after 
thorough drying, a considerable quantity of the acetone-acetylene 
solution can be introduced at about 15 atmospheres pressure and safely 
stored or transported. Cylinders of the acetylene solution are much 
used for automobile and car lighting. Acetylene is considerably 
used for illuminating detached country houses and factories, and for 
the public supply of towns. 

When burned with oxygen, acetylene yields an intensely hot 
flame (2500 C.) Oxy-acetylene blow pipes are much used for autoge- 
nous welding of iron, steel, and other metals ; also for cutting and 
boring steel plates and beams, the metal being rapidly melted in a 
very narrow area at the point of the flame jet. 

Air gas, so-called, is made by blowing a carefully regulated current 
of air through layers of the very volatile petroleum distillates, of from 
80 to 90 Be. The air, carrying sufficient vapor to form a combustible 
mixture, goes directly into the burner, since it cannot be piped very 
far without condensation of the illuminants. Air gas is much used 
where other gas is not available. 

In some parts of Europe, where coal is expensive, gas is made by 
distilling dried peat. The gas contains more carbon dioxide than 
coal gas, and more lime is needed for purification ; it is about 18 
candle power, and considerable tar and ammoniacal liquor are obtained. 
The composition of typical kinds of gas is shown in the following 







Candle power .. . 
Illuminants .... 
Marsh gas .... 
Hydrogen .... 





14.6 Ethane 

Carbonic oxide . 
Nitrogen ... 




1 i 

Oxygen . . .- .r-' ;. 
Carbonic acid 




* C. D. Jenkins. Reports of the Mass. State Gas Inspector. 


The gas produced in by-product coke ovens contains some benzene 
(C 6 H 6 ), and if the gas is not to be used for illuminating purposes, this 
benzene may be washed out of the gas in counter-current scrubbers 
using mineral oil. The oil is then distilled with steam to recover the 
benzene, and the residual oil returned to the scrubber. Coal-gas 
intended to be used for illuminating may be scrubbed in the same 
way, but in this case other illuminants must later be added to the 


Practical Treatise on the Manufacture and Distribution of Coal Gai 

Samuel Clegg, London, 1859. 
Traite theoretique et pratique de la Fabrication du Gaz. E. Borias, Paris 


Manufacture of Gas from Tar, Oil, etc. W. Burns, New York, 1887. 
Fabrikation der Leuchtgase. G. Thenius, Leipzig, 1891. 
The Chemistry of Illuminating Gas. N. H. Humphreys, London, 1891 
Handbuch fur Gas-Beleuchtung. E. Schilling, 1892. 
A Treatise on Gas Manufacture. W. King. (King.) 
Gas Engineer's Handbook. T. Newbiggin, London, 1898. (King.) 
Acetylen in der Technik. F. B. Ahrens, 1899. 
Acetylene. V. B. Lewes, 1900. 
A Textbook of Gas Manufacture. John Hornby, London, 1902. (Bell 

and Sons.) 

Acetylene. F. H. Leeds and W. J. A. Butterfield. London, 1903. 
The Chemistry of Gas Manufacture. W. J. A. Butterfield, 3d ed., 

London, 1904. 

Chemistry of Gas Manufacture. H. M. Royle, 1907. 
Handbook of American Gas Engineering Practice. M. N. Latta, 1907. 
Modern Appliances in Gas Manufacture. F. W. Stevenson. 
Practical Testing of Gas and Gas Meters. C. H. Stone, 1909. 
Gasbeleuchtung und Gasindustrie. H. Strache, Braunschweig, 1913. 
Journal of Gas Lighting. London. Vols. 69, 60, 61, 62, and others. 

(Coal Gas.) 


The tar from the hydraulic main and condensers of the gas works 
is a black, oily, foul-smelling liquid averaging 1.15 sp. gr. Its com- 
position is very complex, and it is mixed with some of the gas liquor, 
retains in solution some constituents of the gas, and carries fine carbon 
in suspension. In the early days of gas making, no use being known 
for tar, it became a great nuisance. But the discovery of important 
derivatives from it has given rise to great industries. 

Coal-tar is used to some extent without treatment for preserving 
timber, as a protective paint and cement in chemical works ; in form- 
ing certain furnace linings ; and as liquid fuel. But much of the tar 
produced is fractionally distilled to separate the more important con- 
stituents. These consist of : (a) the hydrocarbons, the most valuable, 
bodies of a neutral character not affected by dilute acids nor alkali ; 
(6) the phenols, bodies of a weak acid character, and containing 
oxygen ; (c) the bases, containing nitrogen, and often present in such 
small amounts that they are not recovered. The method of distilla- 
tion varies much as the market for the distillates fluctuates, and the 
composition of the tar from different gas works is variable. If ben- 
zenes are low in price, the light oils are collected together. Often the 
phenols are not separated, when the demand for them is not great. 
Some tars are distilled only until the light oils are removed, and the 
residue variously employed. But if anthracene is present, the heavy 
oils are distilled off, and the residue forms pitch. 

When received from the gas works, the tar is run into a tank, or 
cistern, and allowed to stand until the ammoniacal liquor mixed with 
it separates by gravity. To facilitate this the tar may be warmed by a 
steam coil in the tank, especially in cold weather. Gas liquor causes 
frothing in the still, so is removed as completely as possible, and sent 
to the ammonia distiller. 

Formerly, old boilers often served as stills, but in modern works 
the stills are constructed for the purpose. 

In America tar distillation is conducted in a rougher and less per- 
fect manner than in Europe. Horizontal stills of from 15 to 25 tons 
capacity, similar in construction to petroleum stills (p. 339) but smaller, 
are commonly used. The condenser worm is ordinarily wrought iron, 
electrically welded to avoid joints as far as possible. Perforated 
pipes within the still permit the use of steam or compressed air, tc 



assist in distilling the heavy oils and prevent adhesion of coke to 
the plates. 

In Europe vertical cylinders of wrought iron or steel plates from 
three-eighths to one-half inch thick are preferred; the diameter 
is equal to the height ; the bottom is concave, and the top is a cast- 
iron dome, having a manhole, an inlet pipe for the tar, a broad, curved 
vapor pipe (" goose-neck "), and a small overflow pipe (" tell-tale ") ; 
also a thermometer tube and a safety-valve usually ; if the latter is 
omitted, the manhole cover is not screwed down, but closes the open- 
ing by its own weight; should excessive pressure develop within 
the still, the cover is lifted and the vapors escape. The arched bottom 
rises to a considerable height, thus distributing the heat into the 
interior of the mass of tar, and the outlet pipe being placed at the lowest 
point, it is also of assistance in emptying the still. The bottom is 
sometimes protected from direct contact with the flame by a brick 
arch (curtain arch). There is usually a coil in the still, through which 
superheated steam is blown, towards the end of the process, to assist 
in the distillation of the heavy oils. 

These upright stills are set in furnaces so that the flames play 
under the bottoms, and about half-way up the height, through side 
flues in the brick setting. They vary much in size ; in England they 
are from 10 to 20 tons capacity ; in Germany larger ones are used. 

The condenser consists of a cast- or wrought-iron or lead worm, 
placed in a tank of water. A steam pipe is arranged to warm the 
water, if necessary. 

While the still is yet hot from the previous distillation, the tar 
is run in. Since the large mass of cold tar requires some time for 
heating, the fire is started when the still is half full. When the tar 
runs from the tell-tale pipe, the manhole and valves are closed, and 
the heat raised until the contents begin to froth. The overflow 
pipe is then opened, and any ammoniacal liquor which has separated 
is drawn off. The heating is continued carefully until the still-head 
gets warm, and puffs of vapor, and finally drops of liquid, begin to 
come from the condenser. The fire is then moderated, in order to 
prevent boiling over. A closed receiver is placed at the end of the 
condenser, and the distillation is continued very slowly, until the 
temperature reaches 105 C., when, as a rule, the first receiver is 
changed. The distillate is commonly separated as follows : 
First runnings, or " first light oil," to 105 C. 
Light oil to 210 C. 
Carbolic oil, to 240 C, 


Creosote oil, to 270 C. 

Anthracene oil " green oil," above 270 C. 

Sometimes the first runnings and light oil are collected together 
until the temperature reaches 170 C. ; and the distillate between 
170 C. and 230 C. is taken as carbolic oil. The temperature at 
which the distillation is stopped depends upon the quantity of anthra- 
cene in the distillate and upon whether it is desired to produce hard 
or soft pitch. 

The first runnings, or first light oil, contain water, ammonium 
salts, the very volatile oils, and a small quantity of heavier oils car- 
ried over mechanically. After this distillate has run for some time, 
it nearly ceases, although the fire is now increased. This is known 
as the " break," and is the point where the receiver is generally 
changed. During the interval a peculiar sputtering noise (" rattles ") 
is heard in the still, caused by drops of condensed water falling into 
the tar, which is now considerably above 110 C. 

When the liquid begins to run from the condenser again, the 
" second light oil " is collected until the temperature of the tar reaches 
210 C M or until the distillate equals 1.000 sp. gr. This is shown by 
catching some of it in a glass of water; if it forms spherical drops 
which neither sink nor rise in the w r ater, but float at whatever point 
they happen to fall, the receiver should be changed. During this 
period very little cooling water is admitted to the condenser, so that it 
is warmed to 40-50 C. ; the water is then cut off entirely. 

The carbolic oil is distilled until the temperature of the tar 
reaches 240 C., or until a few drops of the distillate cooled on an 
iron plate show crystals of naphthalene. This oil contains phenols, 
and as the naphthalene is less soluble in the heavy oils than in the 
phenols, its crystallization indicates that all the latter have distilled 
off. The warm water in the condenser prevents crystallization in 
the worm ; towards the end of this period it is sometimes necessary 
to heat the water by a steam coil. 

The receiver is again changed, and the " creosote oil " collected 
until the temperature reaches 270 C. The first runnings of this 
contain much naphthalene, but later the quantity present is small, 
and remains dissolved in the heavy oil. This distillate is the least 
valuable and is often not purified further. 

The anthracene oil, or " green oil," collected over 270 C., con- 
tains anthracene, the most valuable constituent of the tar. The 
water in the condenser is now brought to the boiling point. Super- 
heated steam is injected into the hot tar in the still to aid in carry- 


ing over the heavy vapors. The process is generally stopped when 
the distillate becomes " gummy " ; on cooling it has about the con- 
sistency of butter. 

The pitch left in the still is a thick, viscous mass while hot, and 
if run out immediately will take fire in the air. After cooling a few 
hours, it is run out through the pitch cock, and, when cold, hardens 
and is sold as " hard pitch." But the still must be emptied while 
the pitch is warm enough to drain out completely, for if any is left in 
the still the heat radiating from the brickwork will convert it into coke, 
which fastens very firmly to the still bottom and does not dissolve 
when a fresh charge of tar is run in. To facilitate emptying and also 
to supply a demand for "'soft pitch," it is often the practice after the 
anthracene oil is distilled, to pump into the still a certain amount of 
creosote or carbolic oil, or the " dead oils " from which the anthracene 
has been extracted. This mixes with the hot tar and produces a pitch 
of any desired consistency, according to the quantity of oil used. 

Stills are sometimes provided with mechanical stirring apparatus 
to prevent the pitch from burning on the bottom, and to assist in 
mixing it with the oils used for softening it. 

The crude distillates obtained directly from the tar are further 
purified and separated into commercial products. The first runnings 
contain ammoniacal liquor and naphtha, which are usually separated 
by gravity. The former is put with the gas liquor from the tar; 
the latter is usually refined with the light oil distillate. 

The light oil contains benzene, toluene, and xylene, with some 
thiophene, phenols, pyridine bases, and heavy oils. It is distilled in 
stills much like those used for tar, but smaller. Two fractions are 
made, naphtha, which distils under 170 C., being further purified; 
and the last runnings, which are put with the carbolic oil. 

The naphtha is put into a lead-lined vessel provided with an agi- 
tator, and thoroughly mixed with dilute caustic soda solution. This 
combines with the phenols, which are thus removed when the soda so- 
lution is drawn off. After washing the oil with water, about 5 per cent 
of sulphuric acid (sp. gr. 1.83) is added and agitated with the oil, the 
temperature being kept low. This dissolves thiophene, unsaturated 
hydrocarbons and pyridines, and chars and destroys other matter. 
The acid tar thus formed is drawn off and the oil washed several times 
with water, and finally with caustic soda to remove all the acid. The 
washed oil is then redistilled. When collected up to 110 C., the dis~ 
tillate is called " 90 per cent benzol," since that amount by volume 
distils below 100 C. It contains about 70 per cent pure benzene, 


24 per cent toluene, and some xylene. If collected up to 140 C., the 
distillate is known as " 50 per cent benzol," and contains about 46 
per cent pure benzene. Between 140 C. and 170 C. a distillate called 
" solvent naphtha " or " benzine " is obtained. This consists mainly 
of xylenes, cumenes, etc., and is used as a solvent for resins and rubber, 
for thinning paints, and to wash the crude anthracene obtained from 
the anthracene oil. It is also employed to enrich illuminating gas, and 
as a cleansing agent for grease-stained fabrics. It must not be confused 
with petroleum benzine (p. 340), which is of different composition. 

The crude 50 or 90 per cent benzol is chiefly employed in the 
coal-tar dye industry. By careful distillation in a rectifying still, 
such as Coupier's or Sevalle's (p. 12), it yields pure benzene, boil- 
ing at 80-82 C., toluene at 110-112 C., and xylene, 137-143 C. 

The carbolic oil contains phenols and naphthalene ; after cooling, 
the oil is pressed or filtered out of the magma of crude naphthalene 
crystals, which are purified by treating with sulphuric acid and 
heating to destroy the phenol left in them. After separating the 
acid tar and washing, the naphthalene is distilled or sublimed. 

The oil pressed from the naphthalene crystals may be treated by 
either of the following processes to recover the phenols : (a) The oil is 
agitated with dilute caustic soda, which dissolves the phenols, forming 
solution of " sodium carbolate." This separates by gravity from the 
undissolved neutral oils and is drawn off and decomposed with sul- 
phuric acid, or carbon dioxide, or furnace gases, whereby crude carbolic 
acid (phenol) separates as an oily liquid, from which crude phenol crys- 
tallizes on chilling, (b) The oil may be heated with a mixture of lime 
and sodium sulphate, sodium carbolate being formed and calcium sul- 
phate precipitating. After the impurities have settled, the solution of 
phenols (tar acids) is decanted and sold as crude carbolic acid. This 
is purified by repeated distillation in a column apparatus of iron or 
copper, with zinc condensers. Sometimes potassium bichromate and 
sulphuric acid are put into the still to oxidize the impurities. Crystals 
of phenol separate from the distillate on cooling, while the cresols re- 
main liquid. The phenol is separated from all liquid matter by a cen- 
trifugal machine. By treating the alkaline solution of phenols with an 
insufficient quantity of acid, the cresols are precipitated first and may 
be separated, the phenol being separated afterwards with more acid. 

Crystallized phenol, C 6 H 5 OH, melts at 42 C., but the presence of 
a very little water causes the whole mass to liquefy. It boils at 
184 C., and can be distilled unchanged. Carbolic acid is a violent 
poison and has a penetrating odor. It is a powerful antiseptic, 


germicide, and disinfectant. It is the source from which many dyes, 
explosives, and medicinal chemicals are prepared. When dissolved in 
soap, the crude tar acids are often used as antiseptics under the names 
lysol and kreolin ; these are soluble in water or emulsify with it. 

The creosote oil also furnishes naphthalene, which crystallizes on 
cooling. It is filtered, or pressed in presses which have steam-heated 
plates ; the crude naphthalene is washed with caustic soda solution 
and with concentrated acid, and is distilled or sublimed; the oil 
contains cresols and higher phenols, naphthol, and liquid paraffine, 
which have but little value and are not separated. It is chiefly used 
for preserving (" pickling ") timber and railroad sleepers ; the 
timber is thoroughly dried, placed in tanks from which the air is 
exhausted, and the hot creosote oil pumped in under heavy press- 
ure. A small amount is used for lubricant, and as an illuminant 
for outdoor work where smoke is of no consequence. It is also 
used as fuel, and extensively in the preparation of " sheep dips," 
liquids used for destroying ticks and vermin on sheep and cattle. 

Naphthalene is one of the most important constituents of coal-tar, 
forming over 5 per cent of it. It forms shining white platelike 
crystals, which melt at 79 C., and boil at 218 C. It has a peculiar 
penetrating odor, and is much used instead of camphor to protect 
woolen goods and furs from moths ; it is also used to prepare naphthols, 
naphtylamines, and phthalic acids as " intermediates " for the manu- 
facture of dyes. Nitronaphthalene is employed to remove the " bloom " 
from mineral oils (p. 342). 

The anthracene oil, or " green oil," contains about 10 per cent 
anthracene, Ci 4 Hio, together with other solid hydrocarbons, such as 
phenanthrene, chrysen, carbazol, paraffine, and liquid oils of high 
boiling points. The mass is cooled until the solid matter has crystal- 
lized, when the liquid oils are removed by bag filtering or by a filter 
press or centrifugal machine. The crystalline mass so obtained is 
pressed in canvas bags in a hydraulic press at a temperature of 40 C. 
The oils expressed are then again chilled to a low temperature and 
pressed, or are redistilled to recover more anthracene ; then they are 
mixed with the creosote oil or run back into the tar still to soften the 
pitch. The crude 30 per cent anthracene from the press is pulver- 
ized and washed with creosote oil or with solvent naphtha from the 
light oils, which dissolves much of the contaminating substances, but 
does not remove carbazol. The magma is " centriffed " or pressed ; 
the liquid separated is distilled to recover the naphtha, and the residue 
of phenanthrene, having little value, is usually burned for lampblack. 



By these washings, the anthracene is raised to about 50 per cent, 
when it is sold to the alizarine manufacturer. For further purifica- 
tion, it is washed with caustic potash to remove carbazol, and then 
it is sublimed in an atmosphere of superheated steam. It forms 
white plates of pearly lustre, melting at 213 C. and boiling at 360 C. 
It is employed chiefly in the preparation of artificial alizarine. 

The pitch left in the still is either hard or soft, as described on 
page 330. If so soft that it remains liquid when cold, it is often used 
as a black varnish for painting metal work and wood, or for making 
tarred paper or roofing paper. Soft pitch is used as a binder in 
preparing fuel " briquettes " from coal dust. Pitch is also mixed 
with asphalt for making sidewalks and pavements. Soft pitch softens 
at about 38MO C. and melts at 60 C. When a small piece is 
chewed, it coheres together like gum. Hard pitch softens at 75- 
80 C., and melts above 120 C. When chewed, it pulverizes into 
a non-cohesive powder in the mouth. 

The yield of crude products from tars is about as follows : 



Ammoniacal liquor .... 

181 % 

2.3 % 

Light oils . * P" . . 

1 65 


Middle oils (carbolic oil) .... 
Heavy oil (creosote oil) 




Anthracene oil 









The yield of purified products from tar is about as follows : { 
Benzol and toluol . . . 0.22% Cresols . . . . 1.13% 

Xylol and solvent naphtha 0.62 Naphthalene . . 6.40 

Phenol . . . -, . . . 0.40 Anthracene (pure) 0.44 


Das Anthracene und seine Derivate. G. Auerbach, Braunschweig, 1880. 
Die Chemie des Steinkohlentheers. Gustav Schultz, Braunschweig, 1890. 
Die technische Verwerthung des Steinkohlentheers. G. Thenius, Vienna, 


Coal-Tar and Ammonia. Geo. Lunge, 4th ed., London, 1909. 
Die Industrie des Steinkohlentheers und des Ammoniaks. G. Lunge und 

H. Kohler. 2 vols. Braunschweig, 1912. 
Coal Gas Residuals. Frederick Wagner. New York, 1914. 

* Schniewindt, Mineral Industry, 1902, 152. 

t Lunge, Coal-tar and Ammonia, I, 116. 

t Heusler, Chem. Technologic, Leipzig, 1905, 188. 



Petroleum is widely distributed, being found in many places in 
sufficient quantities for profitable working. The principal deposits 
in America are located in Pennsylvania, New York, Ohio, West Vir- 
ginia, Indiana, Illinois, Kansas, Kentucky, Oklahoma, Louisiana, 
Texas, California, Colorado, Mexico, and Canada. The next in im- 
portance to the American oil fields are the Russian, in the Baku dis- 
trict around the Caspian Sea, in the Caucasus mountains, and along 
the northeast coast of the Black Sea. Less important deposits occur 
in Persia, Burmah, Borneo, Galicia, and Roumania. Small deposits 
are worked in Germany, Hungary, Algiers, Japan, Venezuela, New 
Zealand, and in some of the islands of the Pacific. 

Petroleum occurs in all geological formations from the Silurian 
to the Tertiary, the New York and Pennsylvania deposits being in 
the Devonian and Upper Silurian, the Colorado fields in the Creta- 
ceous, and those in California in the Miocene epoch or Middle Ter- 
tiary. The Russian, Galician, and Indian oils are chiefly in the 
Tertiary. In all cases, the strata in which it is found are horizontal 
or but slightly inclined, usually not over 30. It is generally found 
in sandstones or conglomerates, called " oil sands," overlaid with an 
impervious shale or slate. Frequently several layers of sandstone 
are struck, lying between beds of the shale. 

The origin of petroleum has been the subject of much study by 
many eminent chemists. Berthelot regarded it as the product of 
the action of steam and carbon dioxide on the alkali metals. Men- 
deleeff supposed it resulted from the decomposition of metallic car- 
bides by water. This necessitates the acceptance of La Place's 
theory of the formation of the earth, and the assumption that heavy 
metals, such as iron, were among the first substances to condense into 
the liquid and solid state, thus forming the central portion of the 
earth; and that these metals then combined with the carbon from 
the surrounding atmosphere to form carbides, which were after- 
wards decomposed by water, from the cooled surface, which perco- 
lated down through cracks and fissures caused by the cooling and 
shrinkage of the earth's crust. Thus hydrocarbons were formed and 
metallic oxides left in the earth. This theory requires that all petro- 



leums have approximately the same composition, in whatever forma- 
tion they are found, but this is not the case. 

Another hypothesis supposes petroleum to be of organic origin. 
Here again are several theories as to the formation of the oil from 
the vegetable or animal remains. One is that the organic matter, 
probably consisting of vegetable matter and mollusks, decomposed 
under salt water with exclusion of oxygen and at a rather low tem- 
perature.* Another, that only animal matter is the basis of the oil 
and that the nitrogen of the animal tissues escaped as ammonia or 
other nitrogen compounds, and that the remaining fat was subjected 
to a species of dry distillation under great pressure, yielding crude 
petroleum, f There is reason to believe that the New York, Pennsyl- 
vania, and Ohio petroleums are of vegetable origin, { but those of 
California, Texas, and some others contain nitrogen and are found 
in rocks filled with animal remains. 

The crude oil usually consists of hydrocarbons, present in homol- 
ogous series, though oils from different localities show differences 
in these series. The Pennsylvania oils contain members of the marsh 
gas series with the general formulas, C n H 2n +2 ; all of these, from methane, 
CHi, up to solid paraffines with C2yH56, have been isolated from these 
oils. Also, small amounts of the olefine series, C n H2 n , and the ben- 
zene series, C n H 2n -6, and in some oils, sulphur and nitrogen have 
been found. Various crude oils from California, Texas, Oklahoma, and 
Kansas contain asphaltum, as well as paraffine. The Russian oils 
consist largely of the naphthene series, general formula C n H2 n , isomeric 
with the olefines, but differing from them in their properties, so the 
refining is not the same as that of the American oils. 

In many places crude oil comes to the surface in small quantities, 
mixed with the water from springs, the first discoveries having been 
reported as " oil springs." The explorers in central New York, as 
early as 1630, mentioned an Indian remedy containing petroleum. 
Later it was sold as " Seneca oil," by the Seneca Indians. Their 
method of collecting it was to spread blankets on the surface of the 
water on which the oil was floating, wringing it out when the blanket 
became saturated. If the layer of oil was thick enough, it was 
skimmed off with a flat board. 

About the middle of the 19th century, petroleum from various parts 

* Phillips. Am. Chem. Jour,, 16, 409-429. 
f Engler, Ber., 1888, 1816; 1889, 592. 

J Orton. Report on Occurrence of Petroleum, Natural Gas, and Asphalt in 
Western Kentucky. 1891. 

Peckham. Am. Jour. Science, 48. (1894.) 


of the world had begun to attract some attention, and crude methods 
of refining it had been devised; in some few instances this purified 
oil was being used for illuminating. But none of these efforts had 
been very successful, and it was not until 1859, when Mr. Drake 
drilled the first productive oil well near Titusville, Pa., that the real 
development of the petroleum industry began.* The Russian, 
Indian, and Galician oils were mentioned by explorers during and be- 
fore the Middle Ages, but the industries have never been developed 
to any great extent, until within the last thirty years, when the 
Russian fields have become very important. 

The crude oil is obtained by boring tube wells through the shale 
into the sand rock. There is no certainty beforehand that a well 
will yield oil, and, indeed, about one-fifth of those bored in this country 
produced none ; these are called " dry holes." 

The machinery used in oil-well drilling is very ingenious, and a 
great number of special devices have been invented to overcome the 
numerous obstacles encountered. Only the principal tools can be 
mentioned here. The chief one is the " centre-bit " (Fig. 
101), a chisel-shaped piece of steel 4 feet long and weighing 
about 300 Ibs., the cutting edge of which is nearly as wide 
as the diameter of the well. Above the centre-bit 
is the "auger-stem," a rigid bar from 12 to 45 feet 
long, to which the bit is screwed. Its chief purpose 
is to guide the bit and keep the hole straight; it 
also adds weight to the drill. Next above the 
auger-stem is a peculiar piece of apparatus called the " jars " 
(Fig. 102). It consists of two links of steel which have a slid- 
ing motion, one within the other, of from 20 to 24 inches. 
The_ object of this is as follows : the centre-bit frequently 
becomes fastened in the hole, either by fragments of broken 
rock acting as wedges between it and the sides of the well, 
or through sinking into a seam in the rock. Any attempt to 
loosen it by a steady upward pull would break the rope, but 
a sudden upward shock is generally sufficient to loosen it. Fl0 - 102 - 
This is obtained by the movable links of the jars. But they are not 
allowed to close completely, and so give a downward stroke, unless 
the tools become fast in the well. Above the jars is a long, heavy steel 
bar called the " sinker-bar." Through its momentum this gives greater 
effect to the action of the jars. To the top of the sinker-bar the rope 

* A period of wild excitement and speculation followed, the description of which 
by Peckham, Crew, and others, is very interesting reading. 


is attached, by which the entire mass is lifted and dropped, just as a 
pile-driver is operated. The drop allowed for each stroke of the bit 
is about two feet. The rope is fastened to the " temper-screw,'* 
which lowers the tools slightly as the rock is cut away by each blow of 
the bit, and turns them in the hole so that the next cut shall be at a 
slight angle to the last one. When all screwed together, the drilling 
tools form a rod about 60 feet long and weighing about a ton. 

Over the spot where the well is to be drilled a timber or steel struc- 
ture is built, called the " derrick " ; this is from 35 to 80 feet high, 
and from 12 to 15 feet square at the bottom, tapering to about 5 feet 
square at the top. On the floor of the derrick is the windlass for 
handling the drilling tools, the rope passing over a small wheel at the 
top. During the drilling the rope passes through a clutch at the end 
of a large walking-beam, driven by the engine, imparting a rapid up- 
and-down motion to the tools. 

An iron " drive-pipe " is sunk through the drift and clay to the 
solid bed-rock. If the latter is within 15 or 20 feet of the surface, a 
shaft 6 or 8 feet square is sometimes dug down to it. Then the 
drilling of the well proper begins, which is usually 7| inches in diam- 
eter to the bottom of the water-bearing strata.. Then the hole 
is decreased to 5f inches diameter, and a tube, called the " casing," 
is put down ; this is provided with a rubber or leather collar to fit 
closely against the shoulder formed where the diameter of the well 
decreases, making a water-tight joint. Then the hole is continued 
to the oil-bearing strata, by means of a 5j-inch bit. 

At frequent intervals it is necessary to remove the mud and splinters 
of rock. This is done by the " sand-pump," or " bailer," which is a 
long metal tube, having a valve in the bottom. It is lowered until 
a pin on the under side of the valve strikes the bottom of the well. 
The water, which is always present, rushes into the bailer, drawing with 
it the debris ; then the tool is at once raised and the valve closes. 

It is customary to drill some distance into the oil-bearing stratum, 
and sometimes a cavity filled with gas, oil, and water is struck. The 
pressure is occasionally so great as to drive the oil to the surface, some- 
times with great force. Such wells are called " gushers." They seldom 
continue to flow for more than a few days or weeks, when pumping 
must be employed. Some of these gushers have produced enormous 
quantities of oil, as much as 75,000 * barrels a day when at their height. 

But most wells do not gush, and it is now customary to re- 
sort to " torpedoing," in order to increase the yield of oil. A tin 
* Mineral Resources of the United States, 1902, 570. 


shell, from 3 to 5 inches in diameter and from 5 to 20 feet long, is 
filled with nitroglycerine and lowered to the bottom of the well. 
On top of the can is a percussion cap, which is fired by dropping a 
piece of iron, called a " go-devil," weighing several pounds, into the 
well. The resulting explosion cracks and shivers the rock, giving 
the oil a better opportunity to flow into the well. Very often a well 
gushes after torpedoing, and measures are usually taken beforehand to 
dispose of the first heavy rush of oil and water. 

The finished well is prepared for pumping by lowering a 2-inch 
pipe, at the bottom of which is the oil pump, worked by a wooden rod 
inside the pipe. Fig. 103 shows sections through a pumping and 
through a flowing well. In a flowing well no pump 
rod is introduced, but the space between the casing 
and tubing is tightly closed at the top, in order to 
force both gas and oil through the tubing. 

The wells range in depth from 50 to 4000 feet, the 
average in New York and Pennsylvania being from 
1200 to 1800 feet. The cost varies, but from 3000 
to 4000 dollars is about the average. The ordinary 
production varies from one to several hundred barrels 
per day. 

The crude oil is now generally carried from the 
wells to the refineries by pipe-lines, six- or eight- 
inch pipe, through which the oil is pumped. At fre- 

FIG. 103. . . 

quent intervals along the pipe-lines are tanks of from 
30,000 to 40,000 barrels capacity, in which the oil is stored until wanted 
for refining. This system mixes all varieties of oils ; hence, if a special 
kind is required, it must be transported in tank cars or in barrels. 

Crude petroleum is an oily liquid varying in color from greenish 
brown to nearly black ; some varieties are reddish brown or orange 
when viewed by transmitted light. Nearly all show some fluo- 
rescence, and have a rather unpleasant odor. The specific gravity 
varies from 0.782 to above 0.850, in oils from different localities. As 
it comes from the well, more or less gas is dissolved in it, consisting 
chiefly of marsh gas, CH 4 ; ethane, C2H 6 ; propane, CaHg ; and butane, 
C-iHio. A very small amount of phosphorus is often present, but 
seldom more than 0.05 per cent. The oils from Ohio and Canada 
have an unpleasant odor, because they contain some sulphur com- 
pounds. Sand and water are also mixed with the crude oil, but these 
settle on standing in the storage tanks. 

In order to separate the various products from the crude oil, it is 



subjected to fractiohal distillation. The higher the percentage of 
the lighter oils, the more profitable for the refiner ; but many crude 
oils are distilled only for the lubricating oils. A few may be used 
as lubricants without distilling. Considerable petroleum is used for 
fuel, but this is being replaced by the residuum from which the more 
valuable light oils have been separated. 

Refining consists in the separation and purification of the market- 
able products of the crude oil, which is usually separated into about 
five portions. These are naphthas, illuminating oils, lubricating oils, 
paraffines, and coke. The process is usually Worked in two stages : the 
distillation and refining first of the light oils, and then of the heavy oils. 
It is only in the large refineries that both processes are carried out ; usu- 
ally one refiner produces the naphthas, burning oils, and " residuum," 
and another starts with the residuum and finishes the process. 

For distilling the light oils, the cylindrical or horizontal still (Fig. 
104) is used ; this is 30 to 40 feet long by 12 or 15 feet in diameter, 

FIG. 104. 


and is set in a brick furnace, with the upper half of the still exposed 
to the air. It holds 600 to 750 barrels, and is provided with steam 
coils and arrangements for blowing in free steam to carry on the pro- 
cess as a steam distillation if desired. 

The condensers are long straight pipes set in troughs through which 
water flows, or they are coils set in tanks of cold water. They are so 
arranged that the distillates are delivered at some distance from the still, 
to diminish the fire risk. Each pipe is usually provided with a trap by 
which the gases (passing over with the oil vapors) are collected and then 
led under the still and burned, thus economizing fuel. Sometimes the 
veryjight oils are burned with the gases, but they are usually condensed, 
forming the " benzine distillate," or crude naphtha. This is stopped 
when the gravity reaches 62 Be. (sp. gr. 0.729). Then comes the kero- 
sene, or burning oil distillate, until the gravity equals 0.790, or for heavy 
illuminating oils, 0.820. Here the distillation is stopped and the resid- 
uum drawn off, to be distilled for lubricating oils, in the "tar stills." 


At high temperatures oils undergo decomposition ; the heavy oils 
tend to split off gas, hydrocarbons and carbon, forming lighter oils ; 
these reactions are reversible. The process is called "cracking"* 
and the reactions are complex, the number of products formed being 
quite large. The heavy oils decompose into paraffines and olefines of 
lower boiling points and hydrocarbons of the aromatic series may also 
be produced. As an example of what may, perhaps, take place the 
hydrocarbon CigHss (octadecane), boiling at 317 C, may be assumed 
to decompose into Ci H 22 (decane), boiling at 173, and C 7 Hi 6 (heptane), 
boiling at 98, and carbon ; or it may form a paraffine and an define, 
e.g. C 8 Hi 8 (octane), boiling at 125, and Ci H 2 o (decylene), boiling at 
172. The lower boiling product would be put with the naphtha dis- 
tillate, and the higher boiling would form a part of the burning oil. 

During the course of the reactions certain amounts of the aromatic 
hydrocarbons are formed in the order, cymene, xylene, toluene, benzene, 
naphthalene, and anthracene ; these substances are produced progressively 
each by the decomposition of the one preceding, the reactions progressing 
further in the direction indicated, the higher the temperature. Thus the 
formation of aromatic bodies in the distillation of coal is also explained, 
and the conditions indicated under which the formation of the individual 
hydrocarbons will be a maximum. Rittman has succeeded in so controlling 
the temperatures and pressures of the decompositions that the artificial 
manufacture of gasolines and benzenes becomes possible. 

The still, as already shown, has its upper part exposed and thus 
cooled by the air ; a column loosely packed with stones is set above 
the still and the vapors pass into it ; often a dephlegmator is also 
used. The heavy oil vapors partly condense on these cooler surfaces 
and fall back into the boiling residuum, which is much hotter than 
their boiling points, the oils of high molecular weight decompose into 
into bodies of lower boiling points, while some carbon separates and 
forms a coke in the still. The several distillates from the crude oil 
are redistilled and divided into further subdivisions. 

The benzine distillate yieldsf : 

Cymogene, B. P. = 32 F. Sp. Gr. = 0.590-0.610 ] 
Rhigolene, B. P. = 60 F. Sp. Gr. = 0.625-0.631 

Gasoline, B. P. = 115 F. Sp. Gr. = 0.635-0.666 
C Naphtha (Benzine) B. P. = 122-140 F. Sp. Gr. = 0.678-0.700. 
B Naphtha, Sp. Gr. = 0.714-0.718. 

A Naphtha (Petroleum naphtha), Sp. Gr. = 0.741-0.745. 

* The history of the discovery of this process is given in chap, iii of Petroleum 
Distillation, by A. N. Leet, New York, 1884. 

| Boverton Redwood Groves and Thorp's Chemical Technology, Vol. II. 


The burning oil distillate yields : 

110 fire test burning oil (" Standard white "). 1 . 

120 fire test burning oil (" Prime white "). } P 
150 fire test burning oil (" Water white "). 

The residuum from the above distillation is transferred to the 
tar still, or if the distillation has been carried on under vacuum, it is 
known as " reduced oil," and is used to make fine lubricating oils or 

The tar stills are cylindrical, and are set in much the same way 
as those already described, but are encased in brickwork almost to 
the top. They are provided with pipes for introducing superheated 
steam, and are much smaller than the crude oil stills. 

The first distillate is collected until the gravity is about 38 Be. 
(0.834 sp. gr.), and is mixed with the next charge of crude oil, or 
washed with acid and soda and refined for burning oil. Then follow 
several distillates of increasing color and density, which are purified 
as described below, and treated to separate the paraffine wax and 
lubricating oils. The distillation is carried on until the still bottom 
is red-hot, when a gummy yellow distillate, called " yellow wax," is 
collected. This contains anthracene and other hydrocarbons of high 
molecular weight. The residue of coke is valuable for electric light 
carbons and other electrical purposes. 

The fractions collected from the burning oil distillate are more or 
less yellow, colored by tarry matters, which would collect in the 
lamp-wick and soon choke it. To remove these impurities, the oil is 
put into an " agitator," a large, lead-lined, iron tank, where it is 
mixed with from 1 to 2 per cent of concentrated sulphuric acid, and 
the mixture stirred by blowing in air at the bottom. The acid unites 
with the tarry matters, and when the blast is stopped, sinks to the 
bottom and is drawn off as " sludge acid." Water is added, and 
after the mixture is agitated, is drawn off. Next a solution of caus- 
tic soda is introduced (about 1 per cent), and the contents of the tank 
again agitated. Then the oil is again washed with water and drawn 
into the settling tanks, where the suspended water settles out, leaving 
a bright, clear oil. These tanks are very shallow, usually only about 
1 foot or 15 inches deep, but may cover an area of 20 by 30 feet. 
They are exposed to the light and air, and usually contain steam 
coils for warming the oil in winter. 

If the oil is now found to have too low a flash point (p. 344), 
it is run through a " sprayer," an upright pipe with cross-arms of 


small perforated pipe, through which the oil is forced into the air in 
fine jets or spray ; after falling some distance, it is collected in tanks. 
By this exposure to air, any light oils, such as benzine or naphtha, 
are volatilized, and the flash point thus raised. But spraying is less 
frequently necessary now, since more care is taken in the original 

Instead of washing, some kerosenes are redistilled, but this gen- 
erally fails to remove all the yellow color, though, when burned, they 
do not form a crust on the wick, due to traces of caustic soda or sodium 

A small amount of burning oil of very high fire test (about 300 F.) 
is made by treating a crude oil distillate (0.823 to 0.846 sp. gr.) with 
a very large proportion (5 to 7 per cent) of sulphuric acid, washing 
with caustic soda, and redistilling with caustic soda lye in the still. 
This oil is sold as mineral colza, mineral seal, and mineral sperm oil. 

The paraffine oils are treated with acid in agitators which may be 
heated by steam pipes ; they are washed and then chilled and left 
several hours until the paraffine crystallizes. The soft mass is then 
put into canvas bags and pressed at 40 F. in hydraulic presses. 
The crude paraffine cake is again melted, crystallized, and pressed. 
It is then washed with a little benzine and pressed once more. It 
is finally melted and filtered hot through bone-char, or fuller's earth 
and on cooling forms the white commercial paraffine. The oils ex- 
pressed are lubricating oils of various grades. 

After the paraffine is removed, some of the lighter lubricating oils 
are converted into "neutral oils" by bone-char or fuller's earth filtra- 
tion and exposure to sunlight and air, to remove the " bloom," so that 
they may be used to adulterate certain animal and vegetable oils. It 
may also be removed chemically by adding about 1 per cent of nitro- 
naphthalene, or dinitrobenzol, or nitric acid. Bloom has no injurious 
effect upon the oil or machinery. 

Crude petroleums containing sulphur (e.g. those from Ohio and 
Canada) are more difficult to refine, and consequently were formerly 
only used for fuel. Successful methods for refining them are, how- 
ever, now in use. The common process is to pass the vapors from 
the crude oil distillation over copper oxide ; or to collect the distil- 
lates from the crude oil separately and redistil them with a large 
excess of copper oxide, or a mixture of lead and copper oxides in a 
still, which is provided with an agitator. The residue consists of a 
mixture of tar, copper sulphide, and oxide. This is pressed and 
calcined at a low temperature, the combustion of sulphur and tar 


furnishing sufficient heat. The final product is copper oxide, which 
is returned to the process. A solution of litharge in caustic soda is 
sometimes used in the agitator after the usual acid and alkali treat- 
ment, to remove the sulphur, but this is not always a success ; though 
it destroys the offensive odor, traces of sulphur sometimes remain and 
become noticeable on burning. 

The lighter lubricating oils are called " spindle oils " and are 
used on rapid-running bearings. " Machinery oils " form the middle 
grades, and " cylinder oils " are the heaviest. Paraffine in lubricat- 
ing oils is said to reduce its viscosity and cause it to become gummy 
when in use. 

" Reduced oils " are made from the residuum left after distilling 
the burning oils from some crude petroleums by the aid of vacuum 
or by simply exposing certain crude oils to the sun and air in shallow 
tanks which may be gently heated by steam-coils in winter. The 
very light oils soon evaporate and the suspended impurities settle. 
Another process is to let the crude oil flow in thin films over woollen 
blankets suspended in warm rooms ; the very volatile oils evaporate 
and much of the suspended matter is retained by the cloth. By 'these 
methods, oils are obtained which are entirely free from any decomposi- 
tion products due to heating, and from any chemicals such as are used 
in washing and bleaching ordinary lubricating oils. Crude oils of 
high gravity (below 32 Be.) are usually selected for this purpose. 

Reduced oils are valuable lubricators and command a good price. 
Sometimes they are char-filtered to improve their color and quality. 

Vaseline or petrolatum is made from the residuum of vacuum dis- 
tilled crude oils. It is treated with acid and soda, washed and char- 
filtered, and sometimes redistilled in vacuum. 

The Russian petroleums are distilled in much the same way as 
the American, but less acid is used, as the naphthenes are somewhat 
soluble in it. It is found practicable to use continuous stills, as the 
residuum is more fluid than in the case of American oils. The stills 
are heated by separate furnaces and connected in such a manner that 
the overflow pipe from one is the supply pipe for the next, the resid- 
uum from the last passing through coils placed in the supply tank, 
so that the crude oil is warm when it enters the first still. By care- 
ful regulation of the heat and the flow of oil, each still can be made 
to yield a distillate of constant gravity. 

Russian petroleum yields about 38 per cent illuminating oils, 
which is lower than the Pennsylvania oils. Since fuel is scarce, the 
residuum, called astatki, is burned in special burners and furnaces. 



The yield of lubricating oils is large, being nearly 36 per cent. They 
are said to be superior to American lubricators for use in cold countries. 

Oil-testing. The usual test for kerosene is the flame test, i.e. the 
determination of the temperature at which the vapors take fire when mixed 
with air. The point usually taken is the " flash point," the temperature 
at which the oil gives off sufficient vapor to form a momentary flash when 
a small flame is brought near its surface. The " fire test " determines the 
temperature at which the oil gives oft enough vapor to maintain a continu- 
ous flame if ignited ; in other words, it shows the temperature at which the 
oil burns in the air, and is about 20 F. higher than the flash point. Both 
the flash point and the burning point are lower than the boiling point. 

The flash point is determined in a special apparatus, and in many 
states and countries the particular instrument and its dimensions are 
specified by law. In this country, " open testers " are largely used, but 
Abel's " closed tester " has become very popular, and is 
now the legal instrument in some of the states and in 
England and Germany. There are many forms of appa- 
ratus for oil testing, but the two above mentioned cover 
the general principles involved in all. Open testers do 
not represent the conditions prevailing in an ordinary 
lamp ; the closed tester more nearly approaches these, 
and its indication is usually about 20 F. lower than that 
shown by the open tester. 

Tagliabue's open tester (Fig. 105) is very simple. 
A copper water bath (A), heated by the small lamp (B), 
contains the glass dish in which the oil to be tested is 
placed. A delicate thermometer (E) 
is hung to dip into the oil. Some- 
times a stirring apparatus is pro- 
vided for both the water bath and 
the oil. The water bath is slowly 
heated, and at regular intervals of 
temperature a lighted match or small 
gas flame is passed half an inch 
above the surface of the oil. The 
temperature at which a flame passes completely over 
the surface is noted as the flash point. The heating is 
usually continued until the oil catches fire on applying 
the light, when the temperature is taken as the burn- 
ing point. The apparatus is rather crude and is open 
to errors. 

Abel's closed tester (Fig. 106) is more complicated, 
but obviates some of the errors of the open cup. It 
consists of a copper cylinder (K, K) in which is the 
water bath (F). In the upper part of the water bath is 
an air chamber (B) in which is suspended the copper 
vessel (A) carrying the oil. All these vessels are provided with close-fitting 
covers. The cover of (A) has three openings which may be opened or 

FIG. 105. 

FIG. 106. 


closed by a small lever. The cover also carries the thermometer (D), 
dipping into the oil, and a small lamp or gas flame set on an axis at 
(C), so that the flame may be brought directly over the middle open- 
ing in the cover. Usually the lever which moves the cover of the opening 
simultaneously turns the flame down to it. The thermometer (E) dips 
into the water bath, which is heated to 54 C. before the oil is intro- 
duced into (A). When the thermometer (D) registers 18-19 C. the test- 
ing is begun, and repeated with each rise of a degree, until the flash is 
seen. This instrument is officially used in Germany, the lever being 
run by clock-work. It is also used in England, the law requiring a flash 
test of 73 F., which is rather low for safety ; it should not be under 
100 F. In this country, each state has its own standard. Some require 
150 F. fire test in open cups, and others 110 F. Most states fix 110 F. 
flash test. 

Lubricating oils are usually tested for viscosity, gravity, flash, and 
burning points, congealing point, and color. The gravity is usually 
determined with the hydrometer or Westphal balance. In this country, 
the Baume instrument is almost always used. 

Viscosity is determined by relative tests, e.g. the rate of flow of the 
oil through a capillary tube or narrow opening, as compared with the 
rate of flow of pure sperm oil through the same tube or opening. Tem- 
perature is here a very important factor. 

The congealing point, or " cold test," determines the temperature 
at which the oil becomes pasty or solid through the crystallization of 
dissolved paraffine or other matter. This test is of great moment if 
the lubricators are to be used in cold climates. 

Color tests are chiefly made on oils intended for export, by com- 
paring a tube full of the oil with standard glass plates of various tints, 
in a colorimeter. For burning oils the colors range from pale yellow or 
straw to water white. 

Certain animal and vegetable oils, when soaked up in waste, will 
take fire on standing. This is especially true of linseed, cotton-seed, 
corn, lard, and neatsfoot oils, and is caused by the rise in temperature 
due to the oxidation of the oil. If from 40 to 50 per cent of mineral 
oil be added to these oils, this spontaneous combustion is prevented 
to a great extent. This is one of the uses of the neutral oils (p. 342). 


In Scotland, Germany, and a few other countries, mineral oils 
are produced by the destructive distillation of certain bituminous 
shales. These are soft, light brown, or gray rocks, which do not 
contain oil as such, but are permeated with bitumen, a complex 
organic substance similar to pitch. When heated in retorts, this 
decomposes into gas, oily products, ammonia, and tar, leaving a 
carbonaceous residue. The temperature of the distillation greatly 


influences the character of the products, a low temperature affording 
an increased yield of oil. 

The shale is broken to small size and heated to a low red heat in 
vertical retorts into which steam is injected to assist in the distilla- 
tion. Both continuous and intermittent systems of distillation are in 
use, the former being generally employed in Scotland. The shale is 
charged at the top of the retort and when " spent " is drawn while still 
hot upon a grate beneath the retort, where its carbonaceous matter 
(amounting to 10-15 per cent) is burned, thus economizing fuel. 

The products of the distillation pass through a series of pipes 
similar to the hydraulic main and condensers of the coal-gas manu- 
facture. The light naphtha vapors and gas pass into a coke tower 
through which heavy paraffine oil trickles ; this absorbs the naphtha, 
while the gas passes on and is burned under the retorts. In the 
hydraulic main and condensers the other distillates condense in two 
layers, the ammoniacal liquor below and the tar and oil above. These 
are separated by gravity. The ammonia liquor is treated in the same 
way as that from coal gas (p. 151). The oily tar (0.865 sp. gr.)is 
distilled in much the same way as crude petroleum until only solid 
coke remains in the still. The distillates are collected together as 
" once-run oil " and washed in agitators with sulphuric acid and 
caustic soda, and then fractionally distilled. These distillates are 
each purified, yielding commercial naphtha, burning oils, lubricator 
oils, and solid paraffine. 

The acid tar from the washing yields some ammonium sulphate, 
and tarry matter which is used for fuel. The soda tar is treated 
with carbon dioxide, which liberates the creosote, used for the same 
purpose as that from coal-tar. The carbonate of soda solution is 
causticized and used again. 


Ozokerite is a natural, paraffine-like substance containing a small 
quantity of oily matter. It was probably formed by the evaporation 
of petroleum until the more volatile oils had escaped. It occurs in 
irregular seams and masses in the earth in Galicia, in the Caucasus, 
in Utah, and in Colorado. In Galicia it is mined by sinking shafts 
and drifting, following the seams. The wax is separated from the 
earthy impurities by hand picking and by washing, the wax being 
lighter than water and rising to the surface. The residue is boiled 
with water to melt out the remaining wax, which is skimmed from 
the surface. Extraction with benzene is also employed. 


The wax is sometimes distilled, by which light oils, illuminating 
oils, heavy oil, and paraffine are obtained. Or it is refined by treat- 
ing with sulphuric acid and caustic soda, followed by a charcoal or 
bone-black filtration. The product, called ceresine, melts at 61 to 
78 C.* and is similar to beeswax. It appears to belong among the 
olefines, having the general formula C n H 2n . Its color ranges from 
pale yellow to white, according to the degree of refining. 

It is used as candle stock ; for preparing insulating compounds 
for electrical work; in making a black dressing for shoes and har- 
ness leather, and to adulterate beeswax. 


Asphalt or mineral pitch is probably an oxidized residue from the 
evaporation of petroleum. This name is usually applied only to the 
solid bitumens, the semi-solid or liquid bitumen being called maltha, 
or mineral tar. Asphalt generally contains sulphur and nitrogenous 
bodies, but is chiefly composed of hydrocarbons. The crude material 
consists of two chief ingredients, that soluble in petroleum spirit, 
called pctrolene, and an insoluble black substance called asphaltene. 
Asphalt occurs in large quantities in and near the " pitch lake " 
on the island of Trinidad ; also in Cuba, Venezuela, California, Utah, 
Texas, Canada, and in many European countries. The Utah deposit 
is particularly pure (gilsonite) and is much used for black varnish and 
for insulating material. It is also much used as a protective paint for 
the interior of chlorine stills, bleaching powder chambers, acid tanks, 
and for waterproofing purposes. Its chief use is for sidewalks and pave- 
ments, for which it is mixed with pulverized limestone or with the 
natural asphalt rock. The latter falls to a loose granular mass when 
heated until the asphalt softens, and is then rolled and stamped into 
place with hot irons. A certain proportion of purer asphalt, or of the 
heavy petroleum oils, is often added to the mixture to render it more 

Crude asphalt contains much moisture and mineral matter. It 
is refined by heating until melted, whereby the moisture is expelled 
and some of the mineral matter separates by subsidence. Two vari- 
eties of Trinidad asphalt are in commerce, " lake pitch " and " land 
pitch." The latter is harder, and has the higher melting point. As- 
phalt is soluble in carbon disulphide, acetone, and benzene, but not 
in alcohol nor water. When heated, it softens at from 80 to 100 C. 

* Redwood. 



Petroleum Distillation. A. N. Leet, New York, 1884. 

Report on Petroleum ; U. S. Census, 1880. S. F. Peckham, Washington, 


Das Erdol von Baku. C. Engler, Stuttgart, 1887. (Enke.) 
A Practical Treatise on Petroleum. Benj. J. Crew, Phila., 1887. (Baird 


Die deutsche Erdole. C. Engler, Stuttgart, 1888. (Enke.) 
Le Petrole. Henry Deutsch, Paris. 

Das Erdol und seine Verarbeitung. A. Veith, Braunschweig, 1892. 
Petroleum ; Its History, Origin, etc. W. T. Brannt, Phila., 1895. (Baird 

& Co.) 

Die Fabrication der Mineralole. W. Scheithauer, Braunschweig, 1895. 
Treatise on Petroleum, 2 vols. B. Redwood and G. T. Holloway, London, 

1896. (Griffin & Co.) 

U. S. Geological Survey, 8th report. (Formation of Petroleum.) 
Report of Experts on Asphalt Paving. Dept. Public Works, Philadelphia, 


L'Asphalte. Leon Malo, Paris, 1888. (Baudry et Cie.) 
Mineral Oils and their By-Products. I. I. Redwood, London, 1897. 


On the Nature and Origin of Asphalt. C. Richardson, New York, 1898. 
Der Asphalt und seine Anwendung. W. Jeep, Leipzig, 1899. 
The Oil Fields of Russia. A. Beeby Thompson, London, 1904. 
Das Erdol, seine Verarbeitung und Vervendung. R. Kissling, Halle, 

a. S., 1908. 

Lubricating Oils, Fats and Greases. G. H. Hurst, 3rd ed., London, 1911. 
A Short Handbook of Oil Analysis. A. H. Gill, 6th ed., Philadelphia, 

American Chemical Journal, 16, 406. The Origin of Petroleum and of 

Natural Gas. F. C. Phillips. 
Proceedings of the American Academy of Arts and Sciences, Vol. 32. 

Investigations on American Petroleum. Charles F. Mabery. 
Proceedings of the American Philosophical Society, Vol. 36, No. 154. 

Origin and Chemical Composition of Petroleum. S. P. Sadtler. 
Journal of the Society of Chemical Industry : 

1890, 359, The Oil Fields of India, Burmah, etc. B. Redwood. 

1894, 719, Removal of Sulphur from Petroleum. 

1894, 790, Origin of Petroleum. F. C. Phillips. 

1894, 794, Present State of the Petroleum Industry. 

1894, 872, American and Russian Petroleums. 
Journal of the Association of Engineering Societies : 

1894, On the Composition of the Ohio and Canadian Sulphur Petro- 
leum. C. F, Mabery. 
Mineral Resources of the United States, 1882 +. 



These oils are usually called " fatty " oils, to distinguish them 
from the mineral and essential oils. They are very widely dissemi- 
nated in nature, both in plants and in animals, and often form a 
large percentage of the weight of the substance in which they are 
found. They differ from the mineral oils in their chemical composi- 
tion, being compounds of organic acids, with bodies belonging to the 
group called alcohols, i.e. they are esters or compound ethers of the 
organic acids. In the majority of cases, the alcohol from which 
these esters are derived is glycerine, or glycerol, C 3 H 5 (OH)3, a tri- 
atomic alcohol ; but occasionally, e.g. in the waxes, a monatomic 
alcohol is the base. The ethers formed from glycerine with the fatty 
acids are called glycerides, a name which is sometimes applied to the 
oils also. The glycerine radical C 3 H 5 is called glyceryl. 

The acids most commonly found in these glycerides are shown in 
the following tables : 







Butyric . . 

C 4 H 8 2 

- 3 

163 C. 


Caproic . . 

C 6 H 12 O 2 

- 1.5 



Caprylic . . 


+ 15 



Capric . , ." 


+ 30 



Laurie . . 

Cj2H.24O 2 

+ 43.5 


at 100 mm. pressure. 

Mvristic . . 

Ci4H 2 8O2 

+ 54 


at 100 mm. pressure. 

Palmitic . 

CieH 3 2O2 

+ 62 


at 100 mm. pressure. 

Stearic . . 


+ 70.9 


at 100 mm. pressure. 


C2()H4oO 2 

+ 75 

Carnabuic . 


+ 72.5 

C erotic 

C27H 54 O2 

+ 78 






Acrylic ...... 

C 3 H 4 O 2 

8 C. 

140 C. 

Crotonic . . . . . . 

C 4 H 6 O 2 



Hypogaeic j 

Ci6H 30 O 2 


Physetoleic j 
Oleic . . . \. . V . 


Erucic 1 


Brassic] ' * " 









Linoleic .... 
Linolenic .... 
Ricinoleic .... 



Liquid at -18 C. 
-10 C. 


The acids containing ten or fewer carbon atoms in the molecule 
may be distilled under ordinary atmospheric pressure without de- 
composition ; they are called volatile fatty acids. The others given 
in the tables are called non-volatile acids; some of them may be 
distilled undecomposed under reduced pressure or by superheated 

With the exception of a few of the less common oils and waxes, 
only acids having an even number of carbon atoms in the molecule 
occur in the fatty oils. The glycerides composing the greater part 
of the important commercial fats are those of butyric, lauric, pal- 
mitic, stearic, oleic, linoleic, and ricinoleic acids ; to a less extent 
occur the esters of caproic, caprylic, crotonic, and myristic acids. 
The fats are always mixtures of several glycerides, and the propor- 
tion in which these are present determines the nature of the fat, 
whether hard, soft, or liquid ; while certain peculiar properties of 
some fats are due to the presence of one or two particular glycerides. 

The glycerides of palmitic and stearic acids are white crystalline 
solids, melting at 61 and 72 C. respectively; that of oleic acid is 
liquid at ordinary temperature. 

The fatty acids are monobasic, and glycerine being a triatomic 
alcohol, the glycerides are composed of three acid radicals combined 
with one alcohol rest; thus the glyceride of palmitic acid has the 
formula (CieHsiC^s = C 3 H 5 , and is called tripalmitin, or, more often, 
simply palmitin. The glyceride of stearic acid is (Ci 8 H 3 5O 2 )3 = C 3 H 5 , 
called tristearin or stearin. That of oleic acid is (CigHaaC^s EE CsHs, 
called triolein or olein. 

The fats and oils are lighter than water. They cannot be boiled 
or distilled, even under reduced pressure, for when heated much 
above their melting point they decompose. Among other products 
of decomposition is a substance called acrolein CH 2 = CH CHO. 
This is a low boiling liquid, having a very disagreeable odor, and 
whose vapors are very irritating to the eyes. 

Fresh fats are nearly odorless and of neutral reaction, but when 


exposed to the air for some time many of them undergo a change by 
which the glycerides are decomposed and the fatty acids set free, 
while glycerine is formed and usually further decomposed at once. 
This breaking up of an organic ester into free acid and an alcohol is 
called hydrolysis, since the elements of water are taken up by the 
acid and alcohol. Thus if R represent the acid radical, hydrolysis 
of a fat may be represented by the general equation : 

CH 2 OR CH 2 OH 

:HOR + 3 H OH = CHOH + 3 H - OR. 

I I 

CH 2 OR CH 2 OH 

This change is often brought about by the fermentation or putre- 
faction of other substances of a gelatinous or albuminous character 
present in the oil, and is accompanied by numerous secondary re- 
actions, which produce bodies of a very disagreeable odor and taste. 
The oil is then said to be " rancid." 

Hydrolysis may be readily brought about by chemical means, and 
is then called " saponification " ; in this case the reaction is much more 
complete, and these secondary reactions do not occur. The process is 
employed in soap and glycerine manufacture, as will appear later. 

Certain oils are oxidized when exposed to the air, and are con- 
verted into thick gummy or resinous masses, or in thin layers form 
dry, hard, transparent, or translucent films. This change is called 
" drying," and is most noticeable in oils containing the glycerides of 
linoleic, linolenic, and ricinoleic acids, which, being un saturated, 
oxidize very readily. 

The unsaturated compounds of the fatty acid series unite directly 
with hydrogen in the presence of suitable catalyzers, to form saturated 
bodies ; * thus oleic acid (CigH^Cy is converted to stearic acid 
(CigHaeC^), and olein yields stearin, which have greater commercial 
value, owing to their higher melting points. Platinum, palladium, 
copper, nickel, and other metals have been tried as catalyzers, but 
nickel is found most suitable, since it is highly active and of moderate 
cost. The nickel is used as a finely divided, metallic deposit upon 
some kind of inert support, or carrier, as pumice-stone, kieselguhr, 
asbestos, or charcoal. 

A solution of nickel salt is mixed with the pulverized carrier, an 
alkali added to precipitate the hydroxide, and the mass after filter- 
pressing is washed free from soluble matter, and dried. The product 

* Jour. Soc. Chem. Ind., 1912, 1155. 



is ground fine, and the nickel reduced by heating to 300 C., in an 
atmosphere of hydrogen ; precautions to prevent feoxidation during 
cooling and subsequent handling must be taken. 

The reduced nickel catalyzer is then mixed with the oil and intro- 
duced into a vessel where it can be treated at about 175 to 200 C., 
with hydrogen gas, under pressures ranging from atmospheric up to 
25 pounds per square inch. The operation is continued until test- 
portions show that the fat has acquired a sufficiently high melting 
point, when the hot oil is filter-pressed to remove the catalyzer and 
carrier, and then cooled. Oils containing linoleic, linolenic, and other 
less saturated bodies are also converted to hard fats by this treatment, 
but require more prolonged exposure to the hydrogen. 

These hardened fats now find extensive use in the preparation of 
lard substitutes and other food products, and for soaps, and in mak- 
ing lubricants. " Thickened " cotton -seed, peanut, sesame, or other 
edible oils have largely replaced the oleo-stearin from tallow in lard 

Oils and fats are found in every part of plants and animals, certain 
parts being richer than others. In plants, the seeds or fruit generally 
contain the most oil, but the quantity varies, even in the same variety 
of plant, according to the soil, cultivation, climate, and the maturity 
of the fruit. Usually it is in inverse ratio to the amount of sugar and 

starch present. In animals, most of 
the fat is found in the abdominal 
cavity, surrounding the kidneys, or 
in a layer just beneath the skin. 
The latter is especially true in the 
case of marine animals (whales, etc.) 
and those living in cold climates. 

.The vegetable oils are obtained 
by crushing or grinding that part 
of the plant richest in oil, and then 
pressing the crushed material, or ex- 
tracting it with some solvent, such 
as benzine or carbon disulphide. 
Mills for crushing olives are of great 
antiquity, the oldest form being 
light edge-runners of wood or stone, 
FlG - 107 - that did not break the kernels. 

Heavy edge-runners of stone or iron (Fig. 107) are used at the present 
time, but steel rolls and buhr-stone mills are more generally em- 


ployed. The edge-rtinner consists of two heavy rollers (A, A), fixed 
on a common axle (B), and travelling in a circle around a vertical 
shaft (C). The rollers rest on a solid stone or metal bed (D), on 
which the material to be ground is spread. Scrapers (E) are fixed on 
the shaft so that they bring the material directly into the path of the 

The ground pulp is pressed in strong canvas or camel's-hair cloths. 
Sometimes part of the oil is -expressed cold, and the meal is then 
heated and pressed a second time while hot. Cold-pressed oils are 
of lighter color and of better quality, but hot pressing gives a larger 
yield. Wedge-presses and screw- 
presses were used in ancient times, 
but the invention of the hydraulic 
press by Bramah in 1795 revolu- 
tionized oil pressing. Knuckle-joint 
and eccentric presses are later in- 
ventions, but are not so extensively 

The hydraulic press (Fig. 108) 
consists of a large piston or ram (R), 
which is forced out of its cylinder 
(C) by the hydrostatic pressure of a 
liquid pumped into the cylinder in a 
small stream. The bags of pulp (B) 
are placed between the ram and a 
fixed top plate (P), and the oil ex- 
pressed is caught in troughs placed 
around the ram head. 

In 1850 Jesse Fisher of Birmingham, England, invented the extrac- 
tion process, using a volatile solvent such as carbon disulphide, or 
better, petroleum naphtha. The solvent is pumped into a closed 
vessel containing the pulp. After extraction, the solution of oil in 
the solvent is drawn off and the latter recovered by distilling it off 
from the oil. This method gives a larger yield of oil, comparatively 
free from gelatinous matter, but some resins and coloring matter 
may be dissolved, thus contaminating it, and up to the present time 
edible oils are not prepared by this process, owing to the persistence 
of the odor and taste of the solvent. A complicated and expen- 
sive recovery plant which is also costly to operate is required. 
Moreover, if the extraction is carried too far, the residue of crushed 
seed pulp has less value as animal food and is chiefly used as fertilizer 


FIG. 108. 



or fuel. Pressing involves less fire risk and yields a lighter colored 
oil, especially if done cold, while the press-cake from many vegetable 
oils has a high value as cattle food, owing to the oil and proteids 
remaining in it. 

Animal oils are contained in cells composed of membranous 
tissue which putrefies soon after the animal is killed, causing the fat 
to become rancid and have a bad odor. Consequently it must be 
rendered immediately. These oils are obtained by: (a) melting, 
" trying out," or rendering in open kettles. The fat is chopped into 
small bits and heated over a fire with a very little water. The tissue 

shrivels together forming " cracklings," 
which float on the oil and are removed 
by straining and are pressed to obtain 
all the oil. Much care is required to 
prevent overheating, and this process 
has been generally abandoned in favor 
of steam rendering (see below) ; (6) by 
boiling with water to which sulphuric 
acid is sometimes added to decompose 
the cell walls, thus liberating the oil; 
(c) by heating with direct steam under 
pressure in large digesters or autoclaves 
(Fig. 109), breaking down the cell walls. 
The fat is introduced through the man- 
hole (B) which is closed when the di-* 
gester is nearly filled to the top, and 
steam at about 50 pounds pressure is 
admitted by the pipe (C) entering near 
the bottom. Before closing the digester, the fat is sometimes washed 
by flushing with water which runs off by the cocks (D) and (E). 
The foul-smelling gases given off during the rendering are conducted 
away by the pipe (H), and after cooling to condense steam they are 
discharged into the chimney or into a closed sewer. After several 
hours heating, the steam is cut off, the pressure relieved, and the 
digester allowed to remain quiet until the oil has risen to the top, 
leaving the cracklings and condensed water in the bottom of the 
tank. The progress of the separation may be followed by trials at 
the test-cocks (F, F). The water is then drawn off through (E), until 
the oil reaches the level of (G), through which it is then drawn 
off. The cracklings are discharged by dropping the lower manhole 
cover (J). 

FIG. 109. 


In testing fatty oils, certain distinguishing properties and reactions 
are sought. The specific gravity is an important indication as to the 
purity of the sample. It is determined by the Westphal balance, Sprengel 
tube, or specific gravity bottle. 

The saponification value * represents the number of milligrams of 
potassium hydroxide needed to saponify one gram of the oil. It is de- 
termined by saponifying one or two grams of the oil with 25 cubic centi- 
meters of alcoholic potassium hydroxide and titrating the excess alkali 

with hydrochloric acid, using phenolphthalein as indicator. 

The iodine (or bromine) value * represents the percentage of iodine 
(or bromine) absorbed by the oil, forming addition, or to a smaller extent, 
substitution products. The saturated fatty acids and their glycerides 
do not combine with the halogens to any appreciable extent ; but those of 
the oleic or ricinoleic series combine with two atoms of iodine (or bromine) ; 
those of the linoleic unite with four, and of the linolenic with six, atoms 
of the halogen. Thus the determination of this value affords a method 
of determining the percentage of unsaturated fatty acids (or glycerides) 
present in the oil. The weighed amount of oil (0.2 gram) dissolved in 
10 cc. of chloroform is mixed with 30 cc. of a standard solution of iodine 
in mercuric chloride and shaken occasionally during fifteen minutes ; 
15 cc. of potassium iodide solution is added and the excess iodine titrated 

with sodium thiosulphate. f The number of cc. of thiosulphate used, 

multiplied by its value in terms of iodine, gives the number of grams of 
iodine not absorbed by the oil ; the difference between this quantity and 
the amount of iodine added to the oil gives the weight absorbed by the 
oil ; this divided by the weight of oil used and multiplied by 100 gives the 
iodine value. 

The Maumene testj shows the amount of heat developed when oil is 
mixed with sulphuric acid. Fifty grams of the oil are treated with ten 
cubic centimeters of strong acid under exact conditions, and the " rise 
in temperature " observed. 

The elaidin test depends upon the fact that nitrous anhydride (N 2 3 ), 
when brought into contact with olein, converts it into the isomeric solid 
elaidin, but the glycerides of linoleic, linolenic, and isolinolenic acids are 
not affected by this treatment. Thus the non-drying oils become solid, 
while the semi-drying and drying oils remain liquid, or at most, become 
buttery. Five grams of oil are mixed with seven grams of nitric acid 
(1.34 sp. gr.), about one gram of copper wire added, and the glass placed 
in cold water (15 C.) and the oil well stirred. After standing two or three 
hours the solidity of the elaidin cake is examined. 

* Oils, Fats, and Waxes. Benedikt-Lewkowitsch. 

t Oil Analysis. A. H. Gill. 

J Compte Rendu, 35 (1852), 572. 


For convenience in study, the fatty oils are generally classified 
according to certain similarities in their properties and sources. A 
convenient classification is as follows * : 

Oils and Fats. Glycerides. 
Vegetable Oils. 

Drying Oils. (1) 
Semi-Drying Oils. (2) 
Non-Drying Oils. (3) 

Animal Oils. 

Fish Oils. (4) 


Liver Oils. (5) 

Blubber Oils. (6) 
Terrestrial. (7) 


Vegetable Fats. (8) 
Animal Fats. (9) 

Waxes. Non-Glycerides. 



Animal Waxes. (11) 
Vegetable Waxes. (12) 

* Oils, Fats, and Waxes. Benedikt-Lewkowitsch. 



Linseed oil is derived from the seeds of the flax plant, Linum 
usitatissimum, L., which is extensively cultivated in northern Europe, 
Italy, Turkey (near the Black Sea), India, Argentina, and in the 
United States. When the plants are raised for their fibre (p. 490), 
they are pulled up before the seeds are ripe ; such seed must be aged 
several months before pressing, but the best oil is obtained from ripe 
seed. The yield is from 25 to 32 per cent, according as the seeds 
are pressed or extracted. The cold-pressed cake is often heated 
and pressed again. Cold-pressed oil is a clear golden yellow, while 
the hot-pressed product is amber or brown. The latter may be 
" bleached " by treating with a solution of ferrous sulphate and ex- 
posing it to the sunlight. The crude oil is stored until the muci- 
laginous matter and water settle ; the product is called " tanked 
oil." Or the crude oil is refined by agitation with sulphuric acid, 
followed by washing with water. The "tanked" or purified product 
is called "raw oil." 

Press-cake from raw oil is one of the most valuable cattle foods. 

Linseed oil is the most important of the drying oils. It contains * 
about 65 per cent of the glycerides of isolinolenic acid, CisH-soOz, and 
15 per cent each of the glycerides of linoleic, CisH^C^, and linolenic, 
CisHsoC^, acids and 5 per cent of olein. These glycerides absorb 
oxygen, and are converted into an elastic mass, linoxyn, of doubtful 
composition, which has been thought to be insoluble anhydrides of the 
acids. The oil becomes thicker and darker colored, and, when in 
thin films, forms a dry, hard varnish. This drying may be hastened 
by the so-called " boiling " of the raw oil. The latter is heated with 
certain salts (such as litharge, lead acetate, or the peroxide or borate 
of manganese), called " driers." A slight decompositon of the glyc- 
erides occurs, and some acrolein is set free ; also a slight polymeriza- 
tion takes place. Possibly the driers form metallic salts with the 
fatty acids to a small extent, the glycerides being partly saponified 
in the process ; the metallic salts remain dissolved in the oil and act 
as oxygen carriers in the drying, when they are exposed to the air. 
The boiling is carried on in open kettles heated by direct fire or by 

* K. Hazura. Zeit. fur angew. Chem., 1888, 312. 


high-pressure steam, and is sometimes aided by blowing air into the 
hot oil. When the latter has lost from 8 to 10 per cent of its weight, 
the process is stopped. The temperature employed varies with the 
kind of drier used, being highest (250 C.) with litharge; but this 
gives a dark-colored product. The lower the temperature the lighter 
colored the product, and the longer the oil must be heated. By 
heating the oil for several days with borate of manganese at 60 C. 
to 125 C., a very light-colored boiled oil is produced. All boiled oil 
should stand several months, or even a year, before use, in order 
that the impurities may settle. Very little of the drier is dissolved 
by the oil, and the clarified boiled oil is decanted from the residue. 
It dries very readily, and is much used for paint mixing. If the 
boiling is continued for ten or twelve hours, at a high temperature, 
the oil becomes a thick, sticky, viscid mass, used as the basis of 
printers' ink. 

If a small quantity of oil is brought to a high heat with the metal- 
lic salt, a dark-colored liquid " drier " or " japan " is formed, which 
may be mixed with a greater amount of raw oil at a moderate tem- 
perature (100 to 125 C.). This forms a so-called " bung-hole " 
boiled oil, which is lighter colored than if the whole mass of oil had 
been heated to a high temperature. The product is claimed to have 
as good drying properties as the genuine kettle-boiled oil. 

Several grades of linseed oil are in the market, the Calcutta being 
considered the best in this country, while the Western and La Plata 
oils are often of poorer quality. In Europe the Baltic oil * is held in 
high esteem, while the Indian oils are regarded as low grade. Lin- 
seed oil is sometimes adulterated with mineral oil, or with rosin, corn, 
menhaden, or cotton-seed oil. 

Raw linseed oil has a specific gravity of 0.9316 to 0.9354 ; saponi- 
fication value of 189 to 195, and an iodine value of 170 to 188. (Boil- 
ing lowers the iodine number.) It does not yield solid elaidin. It 
is used as a soap stock for soft soap, in some kinds of paint, for varnish 
making, and for rubber substitute. Boiled oil is used for paint, for 
printing inks, for oilcloth making, and in the preparation of linoleum. 
For this last, the partially boiled oil is exposed to the air at a moderate 
temperature (20 to 22 C.), until oxidized to a translucent jelly. 
It is then thoroughly incorporated with ground cork, and is rolled 
into sheets and dried. 

* Lewkowitsch, Oils, Fats, and Waxes. Allen, Commercial Organic Analysis, 
Vol. II. Mcllhiney, Report upon Linseed Oil and its Adulterants, to Commissioner 
of Agriculture of New York State, Albany, 1901. 


By the oxidation of certain oils, as in " drying," considerable 
heat is generated, and if they are exposed in thin layers, on porous, 
inflammable material (e.g. when absorbed in cotton rags or waste), 
spontaneous combustion frequently takes place. This is particularly 
liable to occur with linseed oil ; it may be prevented by the addition 
of mineral oils. 

Hemp oil is obtained from the seeds of the common hemp, Can- 
nabis satim, L. The yield is about 30 per cent. It is a greenish 
yellow oil of 0.925 to 0.930 sp. gr. Its saponification value is 190 to 
191.1, and its iodine value 143 to 148. It is a poor drying oil, but is 
used in paint and as an adulterant for linseed oil ; also as stock for 
soft soap. 

Soja (or soya) bean oil, obtained from the seeds of Soja hispida, 
cultivated in China and Formosa, furnishes an important edible oil, 
and soap stock. Its sp. gr. is 0.9255; saponification value, 193.2; 
iodine number, 137 to 141 ; Maumene test, 86 to 87 ; index of refrac- 
tion, 1.4750 at 20 C. The press-cake is a valuable cattle food. 

Poppy oil is a good drying oil, from the seeds of the poppy, Papaver 
somniferum, L. The yield is about 45 per cent of a thin, yellow, 
odorless oil of 0.924 to 0.937 sp. gr. ; its saponification value is 190 
to 197; iodine value, 134 to 143. It is used as a salad oil and to 
adulterate olive oil ; and in the preparation of colors for artists' use. 

Tung oil, or Chinese wood oil, from the seeds of Aleurites cordata, 
a tree native to China and Japan, is chiefly used for paints, varnishes, 
and in making oilcloth. It is a pale yellow to dark brown in color, 
and dries rapidly, forming a hard film. Its sp. gr. is 0.941 at 15 C. ; 
saponification value, 190 to 1Q7 ; iodine number, 155 to 170. It con- 
tains glycerides of oleic and elseomargaric (CigH&C^) acids. 

Sunflower oil is a pale yellow, palatable, odorless oil, from the 
seeds of the common sunflower, Helianthus annuus, L. The yield is 
about 30 per cent, and the press-cake is a valuable cattle food. The 
oil contains the glycerides of oleic, palmitic, arachidic, and linoleic 
acids. Its sp. gr. is 0.924 to 0.926 ; saponification value, 190 to 194 ; 
iodine value, 120 to 133. It is used as a soap stock, for wool oiling, 
and to adulterate olive oil. 


These oils have an intermediate position between the true drying 
and the non-drying oils, some of them showing distinct drying proper- 
ties, while others do not, as is indicated in their iodine values. 


Corn oil or maize oil is derived from the germ of the common corn, 
Zea Mays, L. The germ (removed from the grain in starch making), 
when pressed, yields a yellow oil of 0.920 to 0.927 sp. gr. Its saponifi- 
cation value is 188 to 193 ; iodine value, 111 to 123 ; Maumene test, 56 
to 88 C. It is used as an edible oil ; in making soap and lubricants ; 
and for rubber substitutes ; the press-cake is an excellent cattle food. 

Cotton-seed oil is derived from the seeds of the cotton plant, 
Gossypium herbaceum, L. After the husks are removed in cylinders 
containing rotary knives, the seeds are crushed in a roller mill. 

The meal, heated in iron kettles at 75 to 90 C., is pressed under 
3000 to 4000 Ibs. per square inch. The yield is about 18 per cent. 
The press-cake is a valuable cattle food, but is mixed before feeding 
with two parts of the seed hulls, straw, or other fodder. 

The crude oil is red or reddish brown in color, and must be re- 
fined for most purposes. After settling, it is pumped into large iron 
tanks having stirring apparatus, and steam coils for heating; here 
the heated oil is agitated for a few minutes with a solution of caustic 
soda of 12 to 18 Be. The alkali combines with the free fatty acid 
of the oil to form a soap, insoluble in the oil. This soap is an effective 
adsorption agent for the coloring and albuminous matter, and sepa- 
. rates together with the excess lye as " foots." The agitator is then 
stopped, the " foots " settle to the bottom, and the clear oil is drawn 
off. The amount of lye, temperature of the oil, and time of agitation 
varies according to the judgment of the operator. The " foots " are 
used for soap stock. The clarified oil is still yellow and for some 
uses is further bleached by treatment with fuller's earth, at a tem- 
perature of about 100 C., and with active stirring for a few minutes ; 
the earth is then filtered out of the oil, leaving it water white or yel- 
lowish color, according to the quality of the oil. On standing or by 
chilling below 12 C., the palmitin and stearin in part crystallize, and 
may be removed by pressing. This solid fat, called " cotton-seed 
stearin," is used in making oleomargarine. The oil expressed is clear 
and light-colored, and is extensively used as a salad oil and to adulter- 
ate olive oil. It is also used in the manufacture of " compound lard," 
" cottolene," etc., for which it is mixed with about one and one-half 
times its weight of beef stearin ; and in butterine and oleomargarine, 
to soften them in cold weather. Cotton-seed oil, hardened by hydro- 
genation, is much used as a substitute for lard as a food product. 

Refined cotton-seed oil has a pale straw color and a specific gravity 
of 0.922 to 0.930. Its saponification value is 191 to 196 ; iodine value, 
101 to 116 ; the elaidin test gives a soft buttery mass ; Maumene test, 


70 to 90 C. It is usually free from acids and has a pleasant taste. 
The poorer grades are used for soap making. It is not often adulterated. 

Sesame or Gingili oil is obtained from the seeds of an East Indian 
plant, Sesamum Indicum, L., which is also grown largely in Egypt 
and Asia Minor. The crushed seeds are first pressed cold and then 
hot. The yield is 30 to 50 per cent of thin, yellow, odorless oil of 
pleasant taste, which does not become rancid on exposure. It con- 
sists of 76 per cent olein, the remainder being glycerides of palmitic, 
stearic, and myristic acids. Its specific gravity is 0.921 to 0.924; 
saponification value, 190 to 194 ; it yields a soft elaidin ; the iodine 
number is 103 to 110 ; and the Maumene test, 65 to 68 C. The best 
quality is used as a table oil or to adulterate olive oil ; the common 
grades are good burning oils or soap stock. 

Rape-seed or colza oil is obtained from the seeds of several vari- 
eties of Brassica campestris, L. The seeds are crushed and heated 
by steam before pressing ; this coagulates the albumin and improves 
the quality of the oil. The yield is about 36 per cent of crude oil 
which is refined by agitation with one per cent of strong sulphuric 
acid and washing with alkali ; this removes traces of sulphuric acid 
and the free fatty acids formed by its action. The lighter colored 
and best grades are generally called colza oil, rape oil being applied 
to the commoner grades. Both contain the glycerides of oleic, stearic, 
and erucic or brassic acids. The specific gravity ranges from 0.922 
to 0.930 at 15.5 C. ; the iodine value is 101 to 117; saponification 
value, 191 to 196 ; by the elaidin test, solidification takes place very 
slowly, frequently requiring 50 to 60 hours, and the elaidin is very 
soft; Maumene test, 70 to 90 C. 

The purified colza oil is a pale yellow and is odorless ; it is chiefly 
used as a condiment and as a burning oil. It is often adulterated with 
hemp, cotton-seed, or fish oils or with rosin oil. Common rape oil is 
used as a lubricant, and being very viscid, is frequently employed as 
a standard for measuring viscosity. When exposed to the air, it 
becomes thick and gummy, but does not really " dry." 

Castor oil is obtained from the seeds of Ricinus communis, L. 
They are cold pressed for the first grade of medicinal oil, and hot 
pressed for the common qualities, about 40 per cent of oil being 
obtained. It is very viscid, of 0.960 to 0.970 sp. gr., and contains 
the glycerides of stearic and ricinoleic acids. Its saponification value 
is 176 to 186; iodine value, 81 to 90; Maumene test, 47 C. It is 
apt to become rancid, and is soluble in alcohol and glacial acetic acid, 
and insoluble in petroleum spirit. Its purgative action is probably 


due to an alkaloid present in it. Large quantities are used in making 
" Turkey-red oil," which is prepared by treating the castor oil with 
sulphuric acid at less than 40 C., and washing with a strong brine 
to remove the excess of acid. The oil is decanted from the brine and 
carefully neutralized with ammonia or soda, by which Turkey-red 
oil, the alkali salt of ricinoleo-sulphuric acid, CisHsspHSOsJOs, is formed. 
Oil thus prepared has largely replaced that made from olive oil for 
use in dyeing cotton with alizarine. Its exact composition is as yet 
uncertain, various views having been advanced.* 

Castor oil is also used for making transparent soaps and common 
soap ; its viscosity being greater than that of any other oil at the ordi- 
nary temperature, it is largely used as a lubricant for heavy machinery. 

By blowing air through hot cotton-seed, linseed, lard, or rape oil, 
it is partially oxidized and converted into a thick viscous oil of very 
high gravity (0.942 to 0.970). Mixed with mineral lubricating oils, 
these " blown oils " are used as substitutes for castor oil for heavy 


These usually contain a high percentage of olein, absorb little or 
no oxygen, and do not dry in the air, yield solid elaidin, and have 
lower iodine values than the drying oils. 

Peanut or earthnut oil is obtained from the fruit of Arachis hypogcea, 
L. The oil is a light greenish yellow, with a peculiar odor and taste, 
but when refined the best quality oil is colorless and has a very faint 
nutty taste. It contains glycerides of arachidic and hypogseic acids, 
besides olein, palmitin, and others. Its specific gravity is 0.916 to 
0.922 ; saponification value, 190 to 196 ; and iodine value, 85 to 105 ; 
Maumene test, 45 to 75 C. It is employed as an adulterant for olive 
oil (formerly also in lard oil), as a salad oil, in butterine, for oiling 
wool, and for soap making. 

Olive oil is obtained from the fruit of the olive tree, Olea Europcea, 
L. Both the fruit pulp and the kernel contain oil, but the former 
yields the better quality. The fruit is crushed in mortars or edge- 
runners (care being taken not to break the kernels) and cold pressed. 
A small quantity of " virgin oil " is thus obtained, which is used as a 
condiment. The residue is stirred up with hot water and pressed 
harder than before; then it is ground a second time, crushing the 

* J. Soc. Chem. Ind., 1883, 537. Liechti and Suida. J. Soc. Chem. Ind., 1884, 
412. Mueller and Jacobs. Dingler's polytechnisches Jour., 254, 346. Schmid. 
J. Soc. Dyers and Colorists, 1891, 69. Scheurer-Kestner. 


seeds, stirred up with hot water, and pressed as hard as possible. The 
final press-cake is extracted with carbon disulphide, or is put into pits 
with water and allowed to ferment for some weeks. The oil rises to 
the top and is skimmed off. 

The several grades of oil obtained are purified by heating to coagu- 
late the albuminous matter, and settling. A dark-colored, mucilagi- 
nous substance, called " foots," deposits, and is used for soap stock. 
The lighter colored oils are used for the table and the others for lubri- 
cators, illuminants, and soap stock. Considerable of the grade called 
" Gallipoli " is used for making " Turkey-red oil " and for oiling wool 
after scouring. 

Olive oils vary in color from pale yellow with a greenish tinge 
(due to traces of chlorophyl) to greenish or brownish yellow in the 
poorer qualities. First-grade oils are odorless and palatable, but 
the lower grades are strong-smelling and usually have a disagreeable 
taste. On exposure to the air olive oil is apt to become rancid. 
The specific gravity varies from 0.914 to 0.918; its saponification 
value is 185 to 203 ; iodine value, 78 to 91.5; the elaidin test shows 
a solid mass within two hours, which is not displaced by inverting 
the vessel ; Maumene test, 41 to 47 C. The oil contains about 72 
per cent of olein and linolein, and about 28 per cent mixed palmitin 
and stearin. Being very expensive, it is frequently adulterated with 
cotton-seed, sesame, or rape-seed oil, while poppy, lard, and peanut 
oils are less commonly used. 


These oils are glycerides, and are liquid at ordinary temperatures. 
They absorb oxygen, do not yield solid elaidin, and have high iodine 
values. The varieties of sperm oil do not belong with this group, 
since they are liquid waxes, although obtained from blubber. 


Fish oils are obtained by rendering and pressing the entire body 
of the fish. The press-cake, consisting of the scales, meat, and bones, 
is ground and utilized as " fish scrap " (p. 166), for fertilizer or for 
feeding swine. 

Menhaden or pogy oil, derived from a small fish, Alosa Menhaden, 
is brownish color, has a fishy odor, and dries in the air. Its specific 
gravity is 0.927 to 0.933 ; saponification value, 189 to 192 ; iodine 
value, 148 to 160 ; Maumene test, 123 to 128 C. It is much used 


in currying (p. 579) ; to adulterate whale oils and linseed oil and as a 
substitute for them. It is itself adulterated with mineral oils. 


These oils contain cholesterol and other biliary ingredients which 
are unsaponifiable. ' 

Cod-liver oil is obtained from the liver of the codfish, Gadus 
morrhua. The livers are rendered by steam heat, and the oil sepa- 
rated, is chilled until the stearin solidifies, when it is pressed and the 
clear oil collected. Three grades are made, pale, light brown, and 
dark brown. The pale oil, used in medicine, is limpid, light yellow, 
having little taste or smell ; its value here may be due to traces of 
biliary substances, making it readily digested and assimilated. The 
darker, less pure grades are used for leather dressing. The oil con- 
tains glycerides of oleic, myristic, palmitic, and stearic acids, some 
volatile fatty acids, and cholesterol ; also traces of iodine and phos- 
phorus. The specific gravity is 0.922 to 0.930 at 15 C. ; saponifica- 
tion value, 182 to 189 ; iodine value, 141 to 159 ; Maumene test, 102 
to 113 C. It is often adulterated with shark-liver oil, seal oil, and 
other fish oils. 

Shark-liver oil is chiefly obtained from the livers of the sunfish, 
Squalus maximus. Its specific gravity is 0.911 to 0.928. It is a clear 
yellow oil, containing a large amount of cholesterol, and is mostly 
used for leather dressing and for adulterating cod-liver oil. 


Whale oil or train oil is obtained from the blubber of the Green- 
land or " right " whale, Balcena mysticetus, and other animals of the 
whale tribe. By boiling the blubber in water, the oil rises to the 
surface and is skimmed off. It is yellowish brown in color and has 
a strong fishy odor. Its composition is variable and but little is 
known about it ; glycerides of some of the lower members of the acetic 
series are often present. The glyceride of valeric acid, C 5 HioO 2 , is 
characteristic of some whale oils. The specific gravity is 0.925 to 
0.930 ; saponification value, 188 to 193 ; iodine value, 120 ; Maumene 
test, 85 to 91 C. Some varieties dry on exposure to the air. Whale 
oil is used for leather dressing, in tempering steel, and as an illuminat- 
ing oil. 

Porpoise oil, derived from the porpoise, Phoccena brachycium, is 
very similar to whale oil, and is obtained in the same way. Its den- 


sity is 0.920 to 0.930; saponification value, 216; it yields a small 
amount of elaidin. The best grades (jaw oil) are used for lubricating 
clocks and watches, the commoner qualities for soap stock, for leather 
dressing, and as illuminating oil. 

Blackfish oil is obtained from the blubber of the blackfish, Globi- 
cephalus melas. It is a pale yellow oil, which separates spermaceti 
(cetyl palmitate) on standing. That from the head and jaw is the 
finest quality, and is used for lubricating clocks and fine machinery. 


These oils have low iodine value and yield solid elaidin. They 
are derived from the feet of cattle, horses, and sheep, or are expressed 
from lard and tallow. 

Neat's-foot oil is made by boiling the feet and shin bones of cattle 
in water. It is a pale yellow, limpid oil of 0.916 sp. gr. at 15 C., is 
nearly odorless, and deposits stearin on standing. Its saponification 
value is 194 ; iodine value, 70 ; it yields a solid or semi-solid elaidin ; 
Maumene test, 47 to 48.5 C. It is nearly pure olein, and does not 
readily become rancid nor gummy when used on machinery. It is 
used for a fine lubricator and for leather dressing. It is often adul- 
terated with fish, rape, cotton-seed, and mineral oils, and other hoof 
oils. Bleached tallow oil is often sold as " neat's-foot." 

Lard oil is prepared by cold pressing lard (p. 367). The best 
quality is limpid and colorless, and consists of olein, with some pal- 
rnitin and stearin, the quantity of these latter depending upon the 
temperature of the pressing; poor grades have a brown color and 
offensive odor. It has a specific gravity of 0.915 at 15.5 C. ; a saponi- 
fication value of 195 to 196 ; iodine value, 56 to 74 ; it yields solid 
elaidin. It is used as an illuminant, as a lubricant, and for oiling wool. 
It is frequently adulterated with cotton-seed oil, cocoanut olein, 
" neutral mineral oil," or rape oil. 

Tallow oil consists mainly of olein, and is obtained by pressing 
tallow (p. 367). It is chiefly mixed with mineral oil for use as a lubri- 
cant. If selected, fresh tallow is rendered at 65 C., and the clear 
oil kept for twenty-four hours in a graining vat, the stearin and part 
of the palmitin crystallize. By pressing, the liquid olein and some 
palmitin is obtained as " oleo oil," which is used for artificial butter 
making. The press-cake (oleo-stearin) is used in making " compound 
lard " (p. 360), and sometimes as a soap or candle stock. Low grades 
of tallow oil are not white, and are called " red oil " in trade ; these 


must not be confounded with the red oil which consists of oleic acid 

(p. 382). 


Palm oil is obtained from the fruit of several varieties of palm, 
Elceis Guineensis, Jacq., native to the west coast of Africa. It is a 
mixture of palmitic acid, palmitin, and olein, and is semi-solid in 
this climate. When fresh, it is red or orange yellow, but on stand- 
ing, especially if exposed to the sunlight, it becomes brownish yellow 
or drab. It may be bleached by heating and blowing in air ; or by 
treating with potassium bichromate and hydrochloric acid. Fresh 
oil has a pleasant odor, but is liable to become rancid, when it con- 
tains a large percentage of fatty acids and has a disagreeable odor. 
Its specific gravity at 99 C. is 0.859 ; the saponification value is 196 
to 202 ; iodine value, 53 to 56. It is used as a candle and soap stock, 
and in making lubricants. 

Palm kernel or palm nut oil is derived from the kernels of the 
fruit of Elceis Guineensis, Jacq. It is similar to and used in the same 
way as cocoanut oil. 

Cocoanut oil is derived from the cocoanut, Cocos nucifera, L. (or 
buty racea, L. f.), the chief commercial supply coming from India, 
Ceylon, and the South Sea Islands. The dried meat (" copra ") of 
the nut is pressed or boiled in water. The oil, which is a solid fat in 
this climate, contains the glycerides of myristic, palmitic, stearic, lauric, 
capric, caprylic, and caproic acids. It melts at 20 to 28 C. ; its sa- 
ponification value is 250 to 268 ; its iodine value, 8.9. It is very liable 
to become rancid. It is much used for soap stock, especially for the 
" cold-process " soaps, and since it is not readily precipitated by salt, 
for marine soaps ; but it needs a strong lye for its saponification. It 
is also said to be used for artificial butter and as a substitute for lard. 
By cold pressing, a solid stearin is obtained which is used in making 

Cacao-butter is obtained from the cacao bean, the seeds of The- 
broma Cacao, L., and is a solid fat having a pleasant odor and the 
flavor of chocolate. It consists of the glycerides of palmitic, stearic, 
and lauric acids, with traces of linoleic and arachidic acids. It is 
used for ointments and salves in pharmacy, and in the manufacture 
of " chocolate creams," and for toilet soaps. It is often adulterated 
with tallow, vegetable oils, beeswax, or paraifine wax. Its specific 
gravity is 0.890 to 0.900 at 15 C. ; saponification value, 192 to 202 ; 
iodine value, 32 to 37.7. 


Japan wax is obtained from a species of Rhus by boiling the fruit 
in water. It is a pale yellow or white, has a greasy feel, and can be 
kneaded in the fingers. It consists of palmitin, CaHXCieHaiC^a, 
with some stearin, and is easily saponified. It is not a true wax. It 
melts at 53 to 54 C., and its specific gravity is 0.970 to 0.980 at 15 
C. It is soluble in benzene, petroleum spirit, and in boiling 97 per 
cent alcohol. It is used for candles, for wax matches, as a furniture 
polish, and for adulterating beeswax. 


Lard is prepared from the fat of the hog. It is rendered at a low 
temperature, and is a softer grease than tallow. It is a mixture of 
palmitin, stearin, and olein. It melts at 28 to 45 C., forming a 
clear liquid. Its specific gravity is about 0.932 ; saponification value, 
195 to 196; iodine value, 59; Maumene test, 24 to 27 C. When 
pure, it is white, nearly odorless and tasteless. By pressing it yields 
lard oil (p. 365). It is often adulterated with water, 25 per cent or 
even more being worked into it; .or with cotton-seed oil and oleo- 
stearin ; or with beef fat and cotton-seed oil. The chief uses of lard 
are for culinary purposes, for soap stock, for butterine, and in oint- 
ments and salves. " Compound lard " is a mixture of oleo-stearin 
and white cotton-seed oil. 

Tallow is the solid fat of the sheep or ox. Before rendering, it 
is customary to break up the tissues by grinding with hollow rolls 
having a rough surface and heated by steam. The rendered tallow 
solidifies at about 34 to 45 C., and is graded according to its ap- 
pearance, hardness, odor, and rancidity. It consists of about two- 
thirds palmitin and stearin, and one-third olein. Its density at 99 
C. is 0.860 to 0.862; saponification value, 195 to 198; iodine value, 
40. It is extensively used for soap and candle stock, for lubricating, 
and as a leather dressing. 

Bone tallow is a soft grease obtained by boiling fresh bones in 
water to extract the marrow and fat. It is dark-colored and foul- 
smelling and usually contains calcium phosphate. It is mainly used 
for cheap colored soaps. 

Butter fat is derived from cows' milk. It is very complex, con- 
taining glycerides of a number of acids of which oleic (60 per cent), 
palmitic, stearic, and butyric (5 per cent) are the most important ; 
small quantities of the glycerides of capric and caproic acids are also 
present. Butter fat has a specific gravity of 0.870 at 99 C. ; its 


saponification value is 221 to 227 ; iodine value, 26 to 35. It is the 
basis of butter, of which it forms about 90 per cent, the remainder 
being water, salt, curds, and coloring matter. It is made by churning 
cream to cause the agglomeration of the fat globules into a solid mass. 
Sour cream churns more easily than sweet cream. The latter is re- 
moved from the milk by a separator * or by skimming before the 
milk sours. Butter from sour cream will not keep unless well salted, 
since it contains sufficient casein to increase its liability to become 
rancid, by which a considerable amount of butyric acid is formed. 
Butter is usually colored with carrot juice, saffron, turmeric, or an- 
nato ; or sometimes with certain coal-tar colors. 

Butterine, oleomargarine, and margarine are butter substitutes 
made from mixtures of animal and vegetable oils, flavored with some 
butter, and colored to imitate it. Oleo oil from tallow, and neutral 
lard are much used. These are mixed with cotton-seed oil in cold 
weather (or with peanut or sesame oil abroad) to increase the per- 
centage of olein. 



Sperm oil is obtained from the blubber and head cavity (" case ") 
of the cachalot, or sperm whale, Physeter macrocephalus, the case 
alone sometimes yielding several barrels of free oil. The composition 
of sperm oil is not definitely known, but it differs materially from 
most oils. It contains no glycerides, consisting mainly of esters of 
monatomic alcohols. Some authorities hold that dodecatyl alcohol, 
Ci 2 H 2 5OH, and its allied homologues, such as cetyl alcohol, CieHss OH, 
are present, but this is denied by Lewkowitsch. The oil holds in 
solution a considerable amount of spermaceti (below), which is usually 
filtered out of the cold oil before it is sold. Sperm oil is a limpid, 
golden yellow liquid, having a slight fishy odor; its specific gravity 
is 0.875 to 0.884 at 15.5 C. ; saponification value, 123 to 147 ; iodine 
value, 81.3 to 85 ; it yields a solid elaidin ; Maumene test, 45 to 47 C. 
It is a valuable lubricator, especially for rapid-running machinery, 
since its viscosity is less than other non-drying fatty oils, and varies 

* Before churning, sweet cream is always allowed to " ripen " ; i.e. to stand a 
few hours undisturbed after separating. Usually a " starter " is added to set up 
lactic fermentation ; by using pure cultures of acid-forming bacteria, the quality 
and flavor of the butter can be much better controlled than when the ripening is 


but little with changes of temperature ; and because it does not be- 
come gummy nor rancid. It is also used for illuminating, for leather 
dressing, and in tempering steel. Because of its high price, it is often 
adulterated with mineral oils or with other fish oils. 

The related Doegling or Bottlenose oil is also a liquid wax. 


Spermaceti is a crystalline wax found in the head of the sperm 
whale and which separates from sperm oil when chilled; it is ob- 
tained by expressing the oil. The brown or yellow scales of crude 
spermaceti are treated with a little caustic potash to remove adher- 
ing oil, and are thus rendered white and translucent while they re- 
tain their crystalline structure. Spermaceti consists mainly of cetyl 
palmitate, Ci 6 H 33 O Ci 6 H 3 iO. It is odorless and tasteless and melts 
at about 45 C. Its specific gravity is 0.943 at 15 C. ; saponification 
value, 108 to 128 ; it burns with a large clear flame. Its chief uses are 
in candle making, in confectionery, and in pharmacy. 

Beeswax is obtained from the honey-comb of bees by melting it 
in hot water; the floating layer of tough brown or yellow wax is 
drawn off into moulds. It may be bleached by exposure in thin 
films to the sun and moist air, or by the moderate action of chromic 
or nitric acid, or hydrogen peroxide. Bleached wax is white, and has 
neither taste nor smell. It consists mainly of myricyl palmitate, 
CaoHeiO CieHsiO, and some cerotic acid, C27H 54 O2. It melts at 63 to 
64 C., and has a specific gravity of 0.965 to 0.969 at 15 C. It is 
often adulterated with water or white mineral powders to increase 
its weight. Stearin, paraffine, cerasin, tallow, and vegetable wax are 
often added as adulterants. It is used in candle making, in phar- 
macy, and for many other purposes in the arts. 

Chinese wax, or insect wax, is secreted by an insect, Coccus ceri- 
ferus, Fabr. The wax is deposited on the branches of certain trees, 
which are cut off and the wax removed by hand. It is melted in 
boiling water to separate the dirt, bark, etc. It is white, crystalline, 
and very hard, without taste or smell. It is soluble in benzine, and 
slightly so in alcohol and ether. It consists of ceryl cerotate, 
C 27 H 5 5O C 27 H53O. Its specific gravity is 0.970 at 15 C., and it melts 
at 82 to 83 C. It is used for fine candles, in medicine, as size for 
paper, and as a furniture polish. 

Wool grease is the greasy substance exuded with the perspiration 
from sheep. It is a complex mixture of alcohols and esters, especially 


of cholesterol and isocholesterol, and palmitic, stearic, myristic, 
carnaubic, and other acids; also potassium salts of these acids. It 
does not contain glycerides and the alcohols appear on analysis along 
with the unsaponifiable matter. 

The grease may be obtained by extracting the raw wool with 
naphtha or other solvent; or the alkaline wash-waters in which the 
wool has been washed may be treated with sulphuric acid, in which 
case the grease also contains fatty acids from the soap used (p. 500). 
It is yellow or brown in color, has an unpleasant odor, and emulsifies 
with water. It is used for leather dressing, and in making axle grease 
and other lubricants. Purified wool grease has a specific gravity of 
0.973 at 15 C. ; iodine value, 25 to 28 ; saponification value, 98 to 102. 

Lanolin is made by washing wool grease with water until all the 
soluble matter is removed, melting by heating in water, skimming 
and allowing it to cool and solidify. Lanolin is much used in phar- 
macy as a basis for salves, ointments, and emulsions. It contains 
about 25 per cent of water, and forms a very soft ointment. 


Carnauba wax is derived from a species of palm, Copernicia ceri- 
fera, Mart., native in Brazil. It forms a coating on the leaves, and 
is removed by shaking or pounding. The raw wax is of a grayish 
or greenish yellow and is very hard, though readily powdered. When 
purified, it has no odor nor taste, melts at 83 to 88 C., and has a 
specific gravity of 0.990 to 0.999 at 15 C. Its constitution is com- 
plex, but it contains myricyl cerotate, CaoHeiO C 2 7H53O, myricyl 
alcohol, CsoHeiOH, cerotic acid, C 27 H54O 2 , and other bodies. It is used 
for candle making and for adulterating beeswax, and in varnish. 


Die Chemie der Austrocknenden Oele. G. J. Mulder, Berlin, 1867. 
Die Fettwaaren und fetten Oele. C. Lichtenberg, Weimar, 1880. 
Die Trocknenden Oelen. L. E. Andes, Braunschweig, 1882. (Vieweg.) 
Technologic der Fette und Oele. C. Schaedler, Berlin, 1883. 
Commercial Organic Analysis. A. H. Allen. Vol. II. London, 1886. 
Das Wachs und seine technische Verwendung. S. Sedna, Wein, 1886. 
Die Fetten Oele des Pflanzen und Thierreiches. G. Bornemann, Weimar, 


Die Untersuchung der Fette, Oele, Wachsarten. C. Schaedler, Leipzig, 1890. 
Les Corps Gras. A. M. Villon, Paris, 1890. 
Les Matieres Grasses. G. Beau visage, Paris, 1891. 

Painters' Colours, Oils, and Varnishes. G. H. Hurst, London, 1892. 
Die Schmiermittel. J. Grossmann, Wiesbaden, 1894. 
Chemical Analysis of Oils, Fats, and Waxes. R. Benedikt. Translated 

by J. Lewkowitsch. London, 1895. 


Chemical Technology. , C. E. Groves and Wm. Thorp. Vol. II. Light- 
ing. Philadelphia, 1895. (P. Blakeston, Son & Co.) 

Animal and Vegetable Fats and Oils. W. T. Brannt, Philadelphia, 1896. 

Oils and Varnishes. J. Cameron, London, 1896. (J. and A. Churchill.) 

Analyse der Fette und Wachsarten. Benedikt u. Ulzer, 3** Auf., Berlin, 

Lubricants, Oils, and Greases. I. Redwood, 1898. 

Oil Chemist's Handbook. E. Hopkins, 1900. 

Vegetable Fats and Oils. L. E. Andes, 2d ed., 1902. 

Oils, Fats, and Waxes. C. R. Alder Wright, 2d ed., London, 1903. 

Cottonseed Products. L. L. Lamborn, New York, 1904. (VanNostrand.) 

Technologic der Fette, Oele, und Wachsarten des Pflanzen und Tierreichs. 
G. Hefter, 4 vols., 1906+. 

Chemie, Analyse und Gewinnung der Oele, Fette, und Wachse. L. Ubbe- 
lohde, 4 vols., 1908+. 

Handbuch der Chemie und Technologie der Oele und Fette. L. Ubbe- 
lohde und F. Goldschmidt, Leipzig. 

A Short Handbook of Oil Analysis. A. H. Gill, 6th ed., Philadelphia, 
1911. (Lippincott Co.) 

Lubricating Oils, Fats, and Greases. G. H. Hurst, 3d ed., London, 1911. 

Chemical Technology and Analysis of Oils, Fats, and Waxes. J. Lew- 
kowitsch, 5th ed., 3 vols., London, 1914. (Macmillan & Co., Ltd.) 


Soaps are metallic salts of certain non-volatile fatty acids, the 
commercial article usually containing a mixture of several of these 
salts. Soaps intended for washing purposes should contain only 
soluble salts of the acids ; i.e. those of sodium, potassium, or ammonium ; 
the calcium, magnesium, lead, and other heavy metal soaps are in- 
soluble in water. 

As already explained, the common fats and oils contain the fatty 
acids in combination with glycerine, forming glycerides, and it is 
from these that soaps are generally made. The process of decom- 
posing the glycerides and forming soap is called saponification, 
although this term is generally used to denote the decomposition of 
any organic ester into its basic alcohol and free acid. Saponification 
is effected in several ways : - 

(1) By the action of water or steam at high temperature or 
pressure : 

3 H 2 = C 3 H 5 (OH) 3 + 3 C 18 H 36 O 2 . 

This hydrolysis may be accomplished at a much lower tempera- 
ture if the water is acidulated with a dilute mineral acid, which serves 
as a catalyzer and accelerates the reaction between the water and the 
glycerides of the fat. The amount needed is small, and it is all found 
unchanged, mixed with the products of the reaction. This method is 
chiefly employed for the preparation of glycerine and to obtain the 
free fatty acid. 

(2) By the action of caustic alkalies : 

C 3 H 5 (Ci 8 H35O 2 ) 3 + 3 NaOH = C 3 H 5 (OH) 3 + 3 Ci 8 H 35 O 2 Na. 

This is the reaction employed in ordinary soap making, the caustic 
uniting with the fatty acid radical to form the soap, i.e. an alkali 
salt of the acid. The glycerine formed is a by-product, and progres- 
sive soap makers have a glycerine recovery plant or sell the lye to a 
glycerine manufacturer. 

(3) By the action of lime; (Milly's process, p. 381). 

The chemistry of saponification was first explained by Chevreul, 
who attributed the cleansing action of soap to free alkali formed by 
the decomposition of the soap when brought into solution. The 
fatty acids are weak and soap solutions are therefore strongly alkaline 


SOAP 373 

by hydrolysis ; the insoluble fatty acids produced by this hydrolysis 
make the solution turbid. Soap removes the dirt by adsorbing on 
the surface of the dirt particles and thus emulsifying it ; in this the 
soap acts as a protective colloid. 

The alkalies commonly used for soap making are caustic potash 
and soda. The former yields a " soft soap," which is liquid under 
ordinary conditions, because of the lower melting point, greater solu- 
bility, and possible deliquescence of potassium soaps. The glycerine 
formed remains mixed in the soft soap. 

Previous to Leblanc's invention of the soda process, soap was 
made with caustic potash derived from wood ashes and lime. Com- 
mon salt was added after the saponification of the fat was complete, 
forming hard sodium soap, according to the reaction : 

KCi8H 35 O 2 + NaCl = KC1 + NaCi8H 35 O 2 . 

But now most soft soaps are made from soda soaps by adding a large 
quantity of water. 

The fatty material (soap stock) varies according to the kind of 
soap desired and the facility with which certain stocks may be 
obtained. For white soaps, the best grades of tallow, tallow-oil, 
palm oil, or cocoanut oil are chiefly used in this country. Cotton- 
seed oil may become rancid and cause yellow or brown spots in 
the product, besides giving it a bad odor and greasy appearance. 
Corn oil is also subject to rancidity. In Europe, Castile soap is 
made from second-quality olive oil, to which some cocoanut oil is 
usually added. 

Laundry soaps are made from tallow, bone grease, and house 
grease, and often palm and cotton-seed oils. Yellow soaps are made 
from these materials, with the addition of a certain proportion of 
rosin. The latter combines readily with alkali, but forms a rather 
soft soap, with good lathering properties ; rosin is cheaper than most 
of the fats, and when used in proper quantities, adds certain valuable 
properties to the soap, and is not an adulterant. 

The non-drying oils, with caustic soda, generally yield the hardest 
soaps, while the semi-drying and drying oils form products of butter- 
like consistency. 

Cocoanut oil saponifies readily with strong lye, without boiling; 
hence is used for " cold-process " soaps. " German mottled," or 
'* olein soaps," are made from crude oleic acid (" red oil "), obtained 
in the candle industry (p. 382). The spent lyes from white or yel- 
low soaps are often used in making red-oil soap, in order to save all 


the alkali, since the oleic acid will combine with the carbonate as well 
as with the caustic. 

Toilet soaps should be made from the best material, but many 
cheap grades are made from poorer stock than laundry soap, and the 
defects covered by high color and perfume. Some toilet soaps are 
made by melting together two or more kinds of soap. 

Good soap cannot be made from poor material. The lye must be 
a caustic liquor, free from other salts, sulphides and sulphites being 
especially injurious, since they cause discoloration of the soap. In 
many large works the lye is prepared by causticizing soda-ash with 
lime. When caustic is purchased, it is simply dissolved to form a 
solution of the desired strength, varying from 18 to 30 Be. 

Soap kettles are square or round, and vary in size from 10 feet in 
diameter by 15 feet deep, to 25 by 35 feet, and capable of holding 
300,000 pounds of soap. In modern factories they are always heated 
by steam ; very small ones, used for remelting toilet soaps, etc., being 

steam-jacketed, and the 
larger ones having both 
open and closed coils. 
A modern form (Fig. 
110) has a conical bot- 

P 1 T *J A J I torn, in which the steam 
coils (A, B) are arranged. 
Such a kettle, calculated 
to hold 100,000 pounds 
of soap, is about 15 feet 
in diameter and 21 feet high, the cone bottom being about 5 feet 
deep, and the cylindrical walls about 16 feet high. It is made of 
f -inch boiler plate, and is sheathed with 2-inch pine staves. It rests 
on stone pillars and foundations, and has large draw-off cocks in the 
cone, for running off waste lyes while the soap is pumped away 
through a pipe (D) passing through the side of the kettle. 

Soaps are made by various processes, but the most common are 
the following : 

(1) The fat is treated with the exact amount of caustic alkali 
needed to saponify it, leaving the glycerine in the soap. The so- 
called " cold-process " soap is the most common example of this 

(2) The fat is boiled with solutions of caustic alkali until saponi- 
fication is complete, or until the soap attains certain desired proper- 
ties. The glycerine remains mixed with the product, as in the case 

SOAP 375 

of soft and " marine " soaps ; or it is excluded, as in the case of yel- 
low, laundry, mottled, and curd soaps. 

(3) A free fatty acid is neutralized by treatment with an alkaline 
hydroxide or carbonate, as in the case of oleic acid. 

The cold process is the simplest of soap-making methods, but 
requires carefully calculated quantities of caustic and fat, and the 
latter must be well refined. Since it is difficult to calculate the exact 
amount of alkali, such soaps usually contain free fat or free alkali, or 
both. Cocoanut oil and tallow are chiefly used, and are melted and 
run into a mixing tank heated by steam, or into a crutcher (p. 376). 
Then a definite quantity of strong caustic soda lye, 32 to 36 Be., is 
added, and the mixture well stirred for a few minutes. The heat of 
the reaction is sufficient to carry it on when once started. After 
saponification is well under, way, the stirring is stopped and the mix- 
ture is run into " frames " (p. 376), where it stands several days, to 
complete the reaction and to cool. This leaves all the glycerine and 
any excess lye in the soap. The product looks well when fresh, but 
is very apt to turn yellow and become rancid. 

Most soaps are boiled. The process is usually divided into several 
stages. The melted fat and lye of about 15 Be. (1.115 sp. gr.) are 
run into the soap kettle together, while free steam is blown in to mix 
them, and to form an emulsion of the oil and lye, which is essential 
to the beginning of saponification or, as the soap-boiler terms it, 
" to kill the stock." When the emulsion forms, the lye has " caught 
the stock." If the lye is too strong at first, it does not " catch," and 
water is added and the heating continued until the emulsion forms. 
Strong lye is then carefully added in small portions at a time, and 
boiling is continued to complete the saponification. If a wooden 
stirring paddle be pushed into the mass at this time, the soap adheres 
to it when drawn out, and long strings of soap hang down from it. 
There is no separation of the lye. When the process is finished, as is 
shown by the soap having a dry, firm feel between the fingers, the soap 
is " grained " or " salted out," by adding common salt. This causes 
a separation of the soap from the lye and glycerine, which is shown 
by the soap sticking to the paddle while the lye runs off. When 
properly salted, the soap boils in broad, smooth patches, and is hard, 
and not sticky, when cold. The steam is then cut off and the soap 
allowed to stand for several hours, when it rises to the top.*' The 
salt lye, which contains most of the glycerine, is drawn off, leaving 
the soap in the kettle. Strong lye, 25 Be. (1.205 sp. gr.) is now added, 
and for yellow laundry soaps, rosin is introduced ; for white soap, 



tallow or cocoanut oil is used instead of the rosin. The boiling is con- 
tinued for two or three days, until the soap becomes clear and semi- 
transparent. This second boiling is called the " rosin change " or 

the " strong change " ; during this 
time, the soap rises fully one-third 
the depth of the kettle, and often 
stands higher than its sides. For 
this reason, the kettle is not filled 
more than two-thirds full at first. 
When the rosin or cocoanut oil is 
saponified, the kettle is allowed to 
stand quietly for a number of hours, 
when the lye is drawn off. The 
next step is called " finishing," 
"settling," "pitching," or "fit- 
ting." Water is added to the boil- 
ing soap until it loses its granular 
appearance, after which it is allowed 
to settle for several days. This 
removes excess caustic and any insoluble impurities. The contents of 
the kettle separate into three layers, the soap on top, and the Ive at 
the bottom, and between them a dark-colored layer, called " nigre," 
containing caustic lye, soap, water, and various organic impurities. 

The lye and nigre are drawn off into separate tanks, and the soap 
is pumped into the crutcher, which is a very efficient mixing machine. 
One form (Fig. Ill) consists of a broad, vertical screw, working within 
a cylinder, which is placed in a 
larger tank. The action of the 
screw draws the liquid soap in at 
the bottom and discharges it over 
the top of the cylinder, to again 
pass through the apparatus. A 
thorough mixing is thus secured. 
The perfume, and any filling mate- 
rial, such as silicate of sodium, 
sodium carbonate, borax, talc, etc., 
are added in the crutcher. These 
ingredients are well mixed with the 
soap, which becomes lighter colored, and then stiff and thick. After 
crutching for from 3 to 15 minutes the soap is run into "frames" 
(Fig. 112), which are large sheet-iron boxes, mounted on wheels, and 

FIG. 112. 



having removable sides. Each frame holds from 1000 to 1700 
pounds, or one crutcher full. When it has solidified, after 24 to 36 
hours, the sides are removed, and the block of soap stands several days 
in the air to cool thoroughly. Then it goes to the " slabber " (Fig. 
113), a machine containing a number of tightly stretched steel wires, 
which are pushed against the block of soap, cutting it into slabs of 
the desired thickness. These then pass through a " cutter," a similar 

Fia. 113. 

machine, which forms them into rough bars, which are put into the 
dry room, kept at a temperature of about 90 F., for 12 to 15 hours. 
They are then run through the press, which forms the commercial bar 
and stamps on it the trade mark, name, or other design. They finally 
pass on an endless belt to the wrappers, who enclose them in separate 
papers and pack them in boxes, which are immediately nailed up for 

" Boiled-down soap " is made by treating the soap, after the lye 
has been drawn off, with strong brine, and then boiling it down. 
Sometimes the soap is settled and the nigre and lye separated before 


boiling down. This reduces the percentage of water in the soap, 
leaving it dry and hard. If soaps in which no rosin is used are boiled 
down on the lye until the latter becomes concentrated enough to pre- 
cipitate the soap, and then run into frames and cooled very slowly, 
the small quantity of lye and other impurities mechanically enclosed 
segregate during the cooling into those parts of the mass which are 
the last to solidify, and cause the appearance called " mottling." By 
adding a small amount of copperas, ultramarine, lampblack, or other 
pigment, the mottling becomes more prominent. Castile or Marseilles 
soaps have a green mottle, changing to red on exposure to the air. 
This is due to the presence of copperas, which precipitates the ferrous 
hydroxide with the lye in the soap ; on contact with the air, the green 
hydroxide is changed to the red ferric salt. Rosin produces a more 
uniform soap, without mottle. 

Toilet soaps are made in the same general way as the yellow soap, 
but from finer stock and with greater care to secure the complete re- 
moval of free alkali. Any excess of alkali is usually carbonated during 
the shaving and milling process. 

Three classes of toilet soaps are made, milled, remelted, and trans- 
parent. Milled soaps are made by shaving thoroughly dried bars of 
good soap to fine chips, and drying again until only about 10 per cent 
water remains. The dried soap is then ground in an edge-runner mill, 
and the perfume or other ingredients desired are added at the same 
time. After thorough incorporation, the soap is forced through a die 
plate by heavy pressure, forming a long bar, which is cut into cakes ; 
these are stamped and pressed into the desired shape. This process 
allows the use of very delicate perfumes and other ingredients which 
would be destroyed by heat. It also furnishes a hard cake which 
does not wear away so rapidly when in use. 

Remelted soaps, chiefly made in England, are prepared by remelt- 
ing one or more kinds of soap, together with the perfumes and other 
ingredients, in a steam- jacketed kettle. By rapid agitation of the 
melted mass with paddles, air bubbles can be disseminated through 
the soap, which gives the cake sufficient buoyancy to float on water 
after stamping. 

Transparent soaps may be made in two ways: (a) A common 
soap is dissolved in alcohol, the solution decanted from insoluble im- 
purities, and the alcohol distilled off, leaving the soap as a transparent 
jelly, which is carefully dried in moulds to form the cake. (6) A 
cold-process soap is made as above and coloring matter, perfumes, etc., 
are added. The glycerine formed, remaining in the soap, causes the 

SOAP 379 

latter to have a translucent appearance. By adding more glycerine, 
with a little alcohol, or a solution of cane sugar, the transparency is 

Special scouring soaps for cleaning metal and unpainted woodwork 
are made by adding powdered sand, glass, or pumice-stone to a yellow 
soap. Strongly alkaline soaps often contain ground soda-ash, borax, 
and sodium silicate as " fillers," or frequently as intentional adulter- 
ants. Sodium silicate is generally added to yellow soaps, as it hard- 
ens them somewhat and possesses detergent properties itself. 

Soap powders are made by mixing soda-ash with a soap solution 
containing just enough water to furnish the crystal water for sal soda. 
Since sal soda does not form above 34 C., the mixture can be made 
hot, and on cooling sets to a dry, solid mass. This is ground, packed, 
and sold as soap powder. Abrasive materials, as powdered quartz 
(silex), or ground feldspar, are frequently added with the soda-ash. 
These soap powders are also made from red oil (p. 382) by neutraliz- 
ing with dry soda-ash. 

A few insoluble soaps of the heavy metals are prepared for use 
in pharmacy, the most important being lead soap or " lead plaster," 
which is made by decomposing a neutral soap with a soluble lead 
salt, or by heating olive oil with a paste of lead oxide in water. 


The materials used for candles are: free fatty acids, especially 
palmitic and stearic ; hydrocarbons, such as paraffine and ozokerite ; 
and certain esters of the fatty acids, especially tallow and waxes. 
The requisites for candle stock are : that it shall burn freely without 
smoke or smell ; that it shall not soften at so low a temperature that 
it loses its form from the heat of its own flame ; and that when melted 
it shall be a fluid capable of being drawn into the wick by capillarity. 
Some glycerides, such as tallow, burn with a foul-smelling, smoky 
flame, and hence are only used in the cheapest candles. Also, they 
soften at too low a temperature, and the candle readily bends and 
gutters. Both these objections also apply to paraffine and to some 
of the solid fatty acids. 

Candles are made by dipping, pouring, and moulding. For 
dipped candles, the wick is repeatedly introduced into the melted 
stock, each layer of fat being allowed to solidify before the next dip. 
Tallow dips, the poorest candle made, are prepared in this way. 

Poured candles are made by pouring the melted stock in a slow 
stream over the wick, which is stretched in a frame. This method 
is used for wax candles, since the wax contracts too much on cooling 
to allow casting. While still plastic, they are rolled on a flat table 
under a board, to give them a uniform diameter. 

Most candles are now moulded in a cylindrical metal form through 
which the wick is drawn in the line of its axis. The mould can be sur- 
rounded with hot or cold water to facilitate the casting and removal 
of the candles. Wicks are of plaited or twisted cotton yarn, usually 
flat, except for tallow dips, when they are round. They are so pre- 
pared that the end curls over and burns off as the candle is consumed, 
thus making snuffing unnecessary ; * also, they are often treated 
with ammonium phosphate or borate to prevent their smouldering 
and emitting bad odors when the candle is extinguished. 

Paraffine, ozokerite, and sperm candles (from spermaceti) are 
moulded. In order to prevent softening at too low a temperature, 
and to render them less brittle when handled, a little stearic acid is 
usually added. 

The most important candle stocks are palmitic and stearic acids 
and paraffine wax. 

* This is accomplished in several ways ; one side of the wick may be dipped in 
size, or one thread be drawn a little tighter than the rest. 



Palmitic and stearic acids are usually made from tallow or palm oil 
by saponifying with lime, or water, the hydrolysis with the latter being 
often assisted by the addition of a little acid. Saponification with 
lime is carried on in two ways : (a) by boiling in open vessels with 
about 16 per cent of lime. The resulting insoluble lime soap consists 
of calcium oleate, palmitate, and stearate. It is separated from the 
lye and free glycerine which is also formed, and is decomposed by treat- 
ment with sulphuric acid and steam, setting free the fatty acid. (6) Or 
the fat may be saponified by Milly's process; i.e., boiled in closed 
vessels called autoclaves, under pressure of from 8 to 10 atmospheres, 
with from 2 to 4 per Cent of lime. The latter probably merely starts 
the hydrolysis, which is finished by the steam and water present. The 
products of the reaction are lime soap, free fatty acid, and glycerine. 
The turbid mixture is treated hot with just sufficient sulphuric acid 
to decompose the lime soap. The calcium precipitates as sulphate, 
while on top of the water (which contains the glycerine) is a layer of 
fatty acid. This is skimmed off and treated with water acidulated 
with sulphuric acid to insure complete decomposition of the lime soap. 

Twitchell's process * is one of the more recent improvements in 
saponification methods. In this a compound of sulphuric acid with 
a fat acid (particularly sulpho-oleic acid), and an aromatic body 
with excess of sulphuric acid, is boiled with the oil or fat and water 
in a tank, until the glycerides are decomposed. The sulpho-fat acid 
is called the "saponifier," and about 1 to 1} per cent is added. The 
mixture is thoroughly stirred and steam blown in to effect the boiling, 
the time of which depends upon the amount of saponifier added; 
with 1 per cent from 12 to 24 hours are required. When saponifica- 
tion is complete, the emulsion is broken by adding sulphuric acid, or 
a mixture of sodium carbonate and sodium sulphate; on settling, 
the fatty acids come to the top and the glycerine lye may be drawn 
off from below. With a neutral fat the addition of some free fatty 
acid is advisable, in order to increase the solubility of the saponifier 
in the fat. The process works at low temperature, the fatty acids 
are of good color, and the yield is good. The fatty acids are much 
used for soap making, as well as for candles ; the glycerine is refined 
in the usual way (p.. 385). 

The melted fatty acids obtained by any of these processes are 
run into shallow pans and allowed to stand a few days at a tempera- 
ture of about 30 C., when the palmitic and stearic acids crystallize. 
The magma is first pressed cold, and then at 40 C., in bags in a hy- 

* Wagner's Jahresbericht, 1900, 548. 


draulic press ; the liquid oleic acid separated forms the commercial 
" red oil " or " olein " employed for soap stock ; the solid fatty acids 
compose the candle stock, which is called " stearin." * It melts at 
52 to 55 C. The yield from tallow or palm oil is 44 to 48 per cent 

Saponification by water alone is accomplished by heating the fat 
in an autoclave with water to about 200 C. under pressure of about 
15 atmospheres. A current of superheated steam is introduced, thus 
thoroughly mixing the contents of the vessel. The free fatty acid 
and the glycerine both distil over with the steam, the former con- 
densing in the first receiver, while the latter passes on to another. 
This process needs much care in the regulation of the heat and to 
secure the complete decomposition of the glycerides, but, when 
properly worked, yields very pure products. The fatty acids are 
chilled and pressed as above described, to separate the olein. The 
yield of stearin is about 50 per cent from tallow or palm oil. Slightly 
rancid stock is more easily decomposed than neutral fat. 

By adding from 4 to 5 per cent of strong sulphuric acid to the 
water in the autoclave, the hydrolysis is accomplished at 120 to 150 
C. A part of the glycerine is converted into glyceryl-sulphuric acid, 

SO 2 <; , while some of the oleic acid, which is an 

\0 - C 3 H B (OH) 2 xCOOH 

unsaturated body, forms sulpho-stearic acid, Ci 7 H 34 \ . By 

X) S0 3 H 

the action of the water, this is converted into hydroxystearic acid, 


si , while sulphuric acid is regenerated. The hvdroxy- 


stearic acid separates as a solid with the free stearic and palmitic 
acids, and in the subsequent purification of these by distillation with 
superheated steam, it is decomposed, separating more water and the 
residue polymerizing to form iso-oleic acid .(CigHjjC^), a solid, melting 
at 45 C. Thus the yield of solid fat acids is slightly increased, being 
about 55 per cent from tallow. 

Another method of acid saponification consists in heating the fat 
with concentrated sulphuric acid for a few minutes only, until the 
cell walls of the fat are destroyed and the hydrolysis is begun. The 
saponification is completed by boiling with water. 

The mixed palmitic, stearic, and oleic acids are chilled and pressed 

* Not to be confounded with the glyceride of stearic acid, p. 350. 


as already described. , Saponification with acid gives a discolored prod- 
uct, which is usually purified by redistilling with superheated steam. 
The liquid " olein " separated from the fatty acids by any saponi- 
fication method is of less value than the solid acids. A process for 
producing palmitic acid from this oleic acid is based on the following 
reaction : 

CigHsA + 2 NaOH = Ci 6 H 3 iO 2 Na + C 2 H 3 O 2 Na + H 2 . 

Caustic soda solution and oleic acid are heated together in an iron 
vessel provided with an agitator, until all the water is evaporated. 
The heat is then raised to a little over 300 C., when the evolution 
of hydrogen becomes active. When the hydrogen ceases to escape, 
the product is treated with water, which dissolves the sodium ace- 
tate and any undecomposed caustic, leaving sodium palmitate undis- 
solved. This is decomposed with sulphuric acid to obtain the free 
fatty acid. But the product is too soft and is an unsatisfactory 
candle stock, hence the method is not now in use.* 

After purification, the free fatty acids, obtained by any of the 
processes above described, are employed for candle stock. The 
aqueous solutions of glycerine (" sweet waters "), resulting from the 
saponification, are used in the manufacture of pure glycerine. 

* J. Soc. Chem. Ind., 1897, 391. 


There are two kind of refined glycerine, CaH^OFOs, on the market, 
dynamite glycerine and chemically pure glycerine. These differ only 
in color and in the content of pure glycerine. It is largely recovered 
from the spent lyes from soap making and from the " sweet waters " 
from the digesters where fats have been saponified with lime or with 
water under pressure.* Much crude candle glycerine is imported 
into this country from Europe, that from France, Italy, and Spain 
being derived from olive oil. 

Spent soap lyes are very dilute solutions of glycerine and con- 
tain much impurity. The successful recovery of glycerine from 
them is one of the recent triumphs of chemical industry. The Van 
Ruymbeke process is most generally used. In this the lye is set- 
tled and drawn off from the sludge. It is then treated with a spe- 
cial chemical called " persulphate of iron," the exact composition 
of which is not disclosed, but which contains about 50 per cent of 
sulphuric acid. It is possibly a mixture of ferrous and ferric sul- 
phates. This forms a copious precipitate, consisting of ferric hydrox- 
ide and iron soaps, which drags down all other insoluble impuri- 
ties. This precipitate is removed by filter-pressing and the clear 
liquid tested for any excess of iron sulphate. If any is present, it 
is exactly neutralized with caustic soda and the precipitate filtered 
off. This leaves the lye almost water white and ready for the evapo- 
ration, which is done under high vacuum (27 to 28 inches), in a still, 
across the middle of which is a steam chest having small vertical 
tubes. Fresh lye is introduced, as the evaporation progresses, to 
maintain the level of the liquid. A salt-catch is placed below the 
steam chest in the evaporator, and in this the salt and sodium sul- 
phate which separate, collect, and at the end of the operation are 
removed through a door in the front. The salt thus recovered is 
sent to the soap maker, to be used again in salting out soap. 

The vapors from the evaporator pass through a series of " catch- 
alls " to retain any lye, and then go to a wet vacuum pump, which is 
provided with a jet condenser. The evaporation is usually carried 
on in two or more stages ; sometimes it is continued to a point at 
which sodium sulphate will crystallize, which is thus removed; by 
further evaporation in vacuum, the common salt is crystallized. 

* Saponification with acid destroys much of the glycerine. 


When the lye attains a density of 32 Be. (1.295 sp. gr.), it con- 
tains about 80 per cent of glycerine, and is called crude glycerine. 
This is then distilled under a very high vacuum (28 to 29 inches) in 
a still consisting of a cylindrical iron shell containing a closed steam 
coil and a perforated pipe, through which superheated steam is intro- 
duced. The glycerine in the crude liquid passes over with the steam 
into coolers, which are simply cast-iron drums, cooled by the out- 
side air. Most of the glycerine condenses here, while the uncon- 
densed steam and some glycerine passes on to a surface condenser. 
The vacuum is maintained by a dry vacuum pump. The glycerine 
collected in the cooling drums is concentrated in vacuum pans until 
its specific gravity reaches 1.262. It is then passed through a filter 
press, which removes any suspended dirt, and gives a clear, bright 
product. As a rule, the glycerine recovered from soap lye is not 
bleached, and is generally sold as dynamite glycerine. 

Chemically pure glycerine is made from candle crude glycerine by 
a modification of the Van Ruymbeke process. The crude liquid, hav- 
ing a density of 28 Be., is diluted, and treated with milk of lime to 
neutralize any acid. It is then treated with a bleaching material known 
as " black," the composition of which is kept secret. After filter-press- 
ing, the glycerine is concentrated to a density of about 31 Be., in an 
apparatus similar to that used for dynamite glycerine. It is then dis- 
tilled, as in the case of the latter, and the product condensing in the 
coolers is thoroughly bleached by treatment with more " black," and 
is then filter-pressed. The density of chemically pure glycerine is not 
required to be so high as that of dynamite glycerine, hence no final 
concentration is necessary. A very fine grade of chemically pure 
glycerine is sometimes prepared from dynamite glycerine by subject- 
ing it to the same process employed for candle crude glycerine. 

Of the other methods for recovering glycerine, only the Glatz 
process needs consideration here. In this the lye is treated with a 
small amount of milk of lime, and then all alkali neutralized with 
hydrochloric acid, and the liquid filter-pressed to remove the precipi- 
tated matter. By evaporating the filtrate under a vacuum, crude 
glycerine is obtained, which is distilled under low vacuum with 
superheated steam, the still being heated by direct fire. The prod- 
uct is then concentrated in a Yaryan or similar evaporator, until 
heavy enough for market. 

The yield of glycerine is always calculated on the amount of fat 
saDonified. By careful work, 6.75 per cent of marketable glycerine 
can be obtained by the Van Ruymbeke process. 


Glycerine is a thick, viscid liquid, having a sweet taste and unc- 
tuous properties. It is soluble in water and in alcohol. High-grade 
dynamite glycerine is of a very pale, yellow color, odorless, and free 
from acids. It contains no iron, lead, or calcium salts, and only a 
trace (0.006 per cent, at most) of chlorides. The ash is not over 
0.01 per cent. The specific gravity should not be less than 1.262 at 
15 C. It is chiefly used in making nitroglycerine; also to some 
extent as a solvent; in the preparation of printers' ink-rolls, and 
for increasing the body or viscosity of other liquids. Chemically 
pure glycerine is colorless, containing less than 0.009 per cent car- 
bonaceous residue, no chlorides, and leaves no ash. Its density is 
about 1.260 sp. gr. It is largely used as a preservative for tobacco; 
for confectionery ; in pharmacy ; in the preparation of cosmetics ; as 
a sweetening agent in fermented drinks, and in preserves ; and owing 
to its non-volatile and non-drying character, as an addition to inks 
intended for rubber stamps. 


Technology of Soap and Candles. R. S. Christiani, Philadelphia, 1881. 

Das Glycerine. S. W. Koppe, Wien, 1883. (Hartleben.) 

The Art of Soap Making. A. Watt, London, 1887. 

Handbuch der Seifenfabrikation. C. Diete, Berlin, 1887. 

Guide pratique du Fabricant de Savons. G. Calmels and E. Saulnier, 

Paris, 1887. 

Traite pratique de Savonnerie. E. Morride, Paris, 1888. 
Manufacture of Soaps and Candles. W. T. Brannt, Philadelphia, 1888. 
Seifenfabrikation. (2 Bander.) A. Englehardt, Wien, 1888. 
Der praktische Seifensieder. H. Fischer, Weimar, 1889. (Voigt.) 
Die Seifen-Fabrikation. F. Wiltner, Wien, 1891. (Hartleben.) 
A Handbook of Modern Explosives. (Glycerine.) M. Eissler, New York, 


Savons et Bougies. J. Lefevre, Paris, 1894. 
Soaps, Candles, Lubricators, and Glycerine. W. L. Carpenter, 2d ed., 

London, 1895. 

Soaps and Candles. J. Cameron, 2d ed., London, 1896. (Churchill.) 
Manufacture of Explosives. (Glycerine.) O. Guttmann, London, 1896. 
Manufacture of Soaps. G. H. Hurst, 1898. 

Soap Manufacture. W. L. Gadd, London, 1899. (Bell & Sons.) 
American Soaps. H. Gathman, 2d ed., 1899. 
Manufacture of Hard and Soft Soaps. A. Watts, 1901. 
Textile Soaps and Oils. G. H. Hurst. 1904. 

American Soaps, Candles, and Glycerine. L. L. Lamborn, New York, 1904. 
Die Gewinnung und Verarbeitung des Glyzerins. B. Lach, Halle, a. S., 

Die Stearinfabrikation. Bela Lach, Halle, a. S., 1908. 
Journal of the Society of Chemical Industry, 1889, 4. O. Hehner. 
American Chemical Journal : 17, 59. Evans and Beach. 
Journal of Analytical and Applied Chemistry : 

IV, 147. J. F. Schnaible. V, 379. E. Twitchell. VI, 423. W. H. Low. 
Railroad and Engineering Journal : 

65, 495 and 551. C. B. Dudley. 67, 199 and 215. C. B. Dudley. 


The essential or volatile oils are liquids which give the peculiar 
odors to plants. They occur already formed in the plants, or are 
produced by the combination of substances in the plant, which react 
in the presence of water. They have strong and characteristic odors 
and pungent taste, and are generally volatile without decomposition. 
They are liquid at ordinary temperatures and are usually nearly color- 
less when fresh, but become darker and thick on exposure. Many 
are optically active. They are nearly insoluble in water, but impart 
their peculiar odor or taste to it. They dissolve in alcohol, carbon 
disulphide, petroleum ether, and fatty oils. Excepting those con- 
taining organic ethers, they are not saponifiable. 

An essential oil is usually composed of several chemical sub- 
stances, all of which are volatile with steam, and may possess either 
open- or closed-chain molecules. A few oils consist almost wholly of 
one constituent. The more important classes of bodies found in 
essential oils are : terpenes of the general formula CioHi 6 ; camphors, 
oxygenated substances of alcoholic or ketone structure; geraniol, 
CioHnOH, and citronellol, CioHi 9 OH, and derivatives of these, for 
the most part of open-chain structure; benzene derivatives, or ring- 
form hydrocarbons, phenols, alcohols, aldehydes, ketones, and acids ; 
aliphatic bodies, consisting of open-chain alcohols, aldehydes, and 
acids, or of esters of these; sulphides, thiocyanates, and nitrogenous 
bodies in a few oils. The oils sometimes contain resins, in solution, 
and are then called oleo-resins, or balsams. 

Some of the essential oils can be prepared synthetically ; some 
are extracted from the plant with solvents, by maceration in fat, or 
by enfleurage, or absorption in fat. But most commercially impor- 
tant essential oils are obtained by distillation with water or steam 
or by pressing. 

In the distillation process, the oil-bearing material is put into a 
still with a considerable quantity of water, which is then brought to. 
boiling. The steam carries the oil into the condenser mechanically, 
where a mixture of oil and water is obtained, which is usually milky 
at first. On standing, it separates into two distinct layers, the oil 
usually, but not always, on top. The water is drawn off, and re- 
turned to the still with the new charge ; or the receiver is so arranged 
that the water returns continuously to the still through a siphon. 



When extraction is employed, alcohol, carbon disulphide, ether, 
or petroleum naphtha may be used. The solvent is evaporated from 
the oil, and recovered. 

Some oils, especially those of lemon and orange, are obtained by 
the use of hydraulic or screw presses. The product is fragrant, but 
rather deeply colored. 

Maceration in fat is employed for some essences which are in- 
jured by high temperatures. The fat used is a perfectly pure and 
sweet lard, tallow, or heavy paraffine oil which is melted in a water 
bath. The flowers or leaves are stirred in and digested until ex- 
hausted. The fat takes up the essential oil and is treated with alco- 
hol, which extracts part of the essence. These alcoholic solutions 
are much used in perfumery; the fat, still containing some of the 
essential oil, is used for pomades and similar purposes. 

Enfleurage is employed for those very delicate oils whose odors 
are destroyed by even moderate heat. The flowers to be extracted 
are Jaid in a wooden frame on the glass bottom of which a thin layer 
of perfectly neutral fat is spread. A number of frames are placed 
in a pile and allowed to stand for some hours, when the flowers are 
replaced by fresh ones. This is repeated until the fat has become 
strongly charged with the perfume. 

Oil of turpentine or spirits of turpentine is derived from conifer- 
ous trees, especially from the pine, Pinus palustris, Mill., and P. 
tceda, L., and from the Scotch fir, P. sylvestris, L. The trees are 
" boxed/' i.e. a cavity is cut near the root, and the bark channelled 
with shallow cuts which lead down to the box. The crude turpen- 
tine (an oleo-resin) flows from the cuts and collects in the box, from 
which it is dipped out at intervals. It forms an exceedingly sticky, 
viscid liquid balsam which is distilled with steam. The volatile oil 
of turpentine (about 17 per cent) passes over with the steam, while 
a residue of resin (rosin or colophony) remains in the still. 

The oils obtained from different varieties of coniferce differ some- 
what in their properties. Three commercial grades are important: 
(a) French turpentine consisting chiefly of a terpene, CioHie, and 
called terebenthene or laewpinene, which has a Isevo-rotary action on 
polarized light rays ; (b) American or English turpentine consisting 
of a terpene, CioHie, called australene, which has the same specific 
gravity, boiling point, and chemical properties as terebenthene, but 
is dextro-rotary ; (c) Russian turpentine which contains the terpene, 
sylvestrine, and some of a pinene resembling australene. The oil first 
distilled is usually washed with caustic soda solution to saponify 


rosin acids, and is then redistilled for " rectified spirits of turpen- 
tine." Commercial oil of turpentine or " turps " is a water-white, 
mobile, refractive liquid of 0.640 to 0.872 sp. gr., distilling between 
156 and 170 C. It is insoluble in water and in glycerol, but solu- 
ble in ether, absolute alcohol, carbon disulphide, chloroform, benzene, 
fatty and essential oils. It dissolves sulphur, phosphorus, wax, caout- 
chouc, and resins, and is used as a solvent in varnishes and paints. 
It burns with a smoky flame. It absorbs oxygen from the air, be- 
coming resinous. According to Kingzett, oxidation of turpentine 
forms camphoric peroxide, CioHi 4 O 4 , which with water yields cam- 
phoric acid and hydrogen peroxide. By passing air into Russian 
turpentine in the presence of warm water, the disinfectant " sanitas " 
is made. 

Turpentine is now largely produced by the destructive distilla- 
tion of resinous pine wood, often with the aid of steam injected into 
the retort ; acetic acid and wood alcohol are by-products. 

Camphor,* doHi 6 O, is an oxygenated essential body (probably a 
ketone) occurring in some crude volatile oils. Commercially it is 
obtained from the wood of the camphor laurel, Cinnamomum Cam- 
phor a, Nees & Eberm., native in Japan and Borneo. The trunk and 
branches of the tree are roughly distilled with water, and the crude 
camphor purified by sublimation. 

Artificial camphor may be made in several ways,f by oxidizing 
borneol or isoborneol with permanganate, ozone, oxygen, air, chlorine, 
or nitrous gases. Catalytic reagents may be used to accelerate the 
reaction, as when the vapors of isoborneol and air or oxygen are 
passed over platinized asbestos, metallic copper, or bits of earthen- 
ware at 175 to 180 C., thus producing a mixture of camphor, cam- 
phene, and isoborneol, from which the camphor is separated. Or 
isoborneol dissolved in benzene is treated with chlorine ; camphor is 
produced and remains dissolved in the benzene from which it is 
crystallized. The reaction is CioHisO + 2 Cl = 2 HC1 + Ci Hi 6 O. 

Isoborneol Camphor 

Camphor is a white, translucent body having a penetrating odor 
and pungent taste ; it melts at 175 C., boils at 204 C., is volatile at 
ordinary temperatures, and burns with a luminous smoky flame. Its 
specific gravity is 0.986 to 0.996. It is slightly soluble in water, 
easily so in alcohol, ether, chloroform, carbon disulphide, acetone, and 
essential oils. It is largely used in the manufacture of celluloid 

* J. Soc. Chem. Ind., 1884 (3), 353. 

t Ibid., 1904, 75, 881 ; 1905, 249, 857, 902, 1188. U. S. Pats., 770940, 790601, 
801483, 801485, 802792, 802793. 


(p. 584), in explosives, in medicine, and pharmacy, and as a protec- 
tive against the ravages of insects. 

Thymol, CioHi 3 OH, is a phenol occurring in the oil of thyme 
and in some other volatile oils. It is similar to carbolic acid in its 
character, and it is obtained by washing the crude oil with caustic 
soda, the alkaline solution of thymol being separated and decom- 
posed with mineral acid ; or the oil is chilled and the thymol crys- 
tallizes and may be filtered out. 

It is a colorless crystalline body, having a specific gravity of 
1.028, and melting at 44 C. It is very slightly soluble in water, 
but readily so in alcohol, glacial acetic acid, ether, etc. It is a pow- 
erful antiseptic, and is much used in medicine and in pharmacy. 

Menthol, CioHig OH, is an alcohol occurring in oil of pepper- 
mint, and which crystallizes when the oil is chilled. It is a white 
solid, very sparingly soluble in water, but readily so in ether, alco- 
hol, and fixed and volatile oils. It does not combine with caustic 
alkalies. It melts at 41 to 43 C. It is much used as a remedy for 
neuralgic pains and headache. 

The essential oil of almonds is produced by the action of emulsin, 
a nitrogenous ferment upon amygdalin, a glucoside. To obtain it, 
the marc of almond kernels left after pressing for the fixed oil is 
distilled with water. It contains benzaldehyde, with some hydrocyanic 
acid and other nitrils. It is purified by redistillation over a mixture 
of lime and ferrous sulphate. It is readily oxidized on exposure to 
the air, forming benzoic acid. Artificial almond essence is made by 
boiling benzal chloride with lead nitrate or calcium carbonate and 
water. This oil is used in making dyes and as a flavoring extract. 

Nitrobenzene is used under the name " mirbane" as a substitute 
for almond essence for scenting soaps. 

Oil of bergamot is prepared from the fruit of a species of orange, 
Citrus Bergamia, Risso, by hand pressing or distillation with water. 
It is a light green, pleasant-smelling oil, containing a large amount 
of a terpene, citrene, CioHie, boiling at 175 to 177 C. It is chiefly 
used in perfumery. 

Oil of Cajaput, prepared from the leaves of Melaleuca Leucaden- 
dron, L., is a green liquid of peculiar odor, distilling at 170 to 180 C. 

Cedar oil is obtained by distilling the wood of red cedar, Juni- 
perus Virginiana, L., with water. It contains a mixture of cedrene, 
CisHjn, and a camphor-like body, Ci 5 H 2 6O. 

Chamomile oil, distilled from Anthemis nobilis, L., consists of 
isobutyl and amyl esters of angelica and tiglic acids. 


Cinnamon oil or oil of cassia is distilled from the inner bark of 
Cinnamomum Zeylanicum, Nees. It is a yellow oil, consisting mainly 
of cinnamic aldehyde, with a little cinnamic acid. It is slightly 
heavier than water. 

Oil of cloves is obtained by distilling cloves (the flower buds of 
Eugenia caryophyllata, Thunb.) with water. It is a mixture of a ter- 
pene, Ci 5 H 2 4 (boiling at 251 C.), and eugenol, CioHi 2 O 2 . It is yellow, 
of a penetrating odor, and heavier than water. 

Eucalyptus oil, distilled from the leaves of several Australian 
trees, Eucalyptus Globulus, Labill., and others, is used in perfumery, 
in medicine, and in scenting soaps. It contains terpenes (especially 
pinene, doHi 6 ), cymene, and eucalyptol or cineol, CioHi 8 O. 

Geranium oil is distilled from the leaves of Pelargonium Radula, 
L'Herit. Its odor resembles that of rose oil, which it is chiefly used 
to adulterate. 

Lavender oil is distilled from the flowers of Lavandula vera, D. C. 
It has little odor when first prepared, the perfume being developed 
by exposure to the air. Oil of spike is obtained from L. Spica, Cav. 
It is similar to lavender oil and is used in porcelain painting. 

Oil of lemon is expressed from the rind of the fruit of Citrus Li- 
monum, Risso. Poor grades are made by distilling the rind. The 
oil contains a terpene (limonene), CioHie, boiling at 176 C. It is 
chiefly used in perfumery, and as a flavoring essence in confectionery. 

Mustard oil is distilled from the seeds of Brassica nigra, Koch., 
after the fixed oil has been removed by pressing. It contains nitro- 
gen and sulphur, and its essential principle is allyl thiocarbamide, 
CsH 5 N : CS. It is a pale yellow oil of 1.015 to 1.025 sp. gr., boiling 
at 148 C., and having a pungent, disagreeable odor. It is a power- 
ful irritant and produces blisters on the skin. It is not present in 
dry seeds, but is formed by the action of a ferment, myrosin, upon a 
glucoside, potassium myronate, in the presence of water. Artificial 
mustard oil is prepared by distilling allyl iodide with potassium thio- 
anate : 

C 3 H 5 I + KSNC = KI + C 3 H 5 N : CS. 

il of peppermint, obtained by distilling the herb Mentha piperita, 
L., is a colorless or greenish yellow liquid, of strong pungent taste 
and odor, having a specific gravity of 0.900 to 0.920. It is a mix- 
ture of menthol, Ci Hi 9 OH, with several terpenes. It is much 
used in medicine and as a flavoring essence. 

Attar of roses is obtained by distilling the flowers of various species 
of rose. It is a pale yellowish liquid, somewhat lighter than water, hav- 


ing a very delicate, rich odor. It crystallizes at ordinary temperatures 
and deposits an inodorous body resembling paraffine. The constitu- 
tion of the oil is not known. Owing to its high price, it is frequently 
adulterated with geranium oil, which resembles it somewhat in odor. 

Oil of rue is distilled from the herb Ruta graveolens, L. It con- 
sists mainly of methyl nonylketone, CgHi 9 CO CHs. 

Oil of sassafras is distilled from the root of Sassafras officinale, 
Nees & Ebern. It contains safrol, Ci Hi O 2 , and some pinene, Ci Hi 6 . 
Safrol melts at 8 C. and boils at 228 to 235 C. Sassafras oil is 
much used for flavoring. 

Oil of thyme or origanum is derived from the leaves and flowers 
of Thymus vulgaris, L. It is yellowish red, has a pungent taste, 
and a specific gravity of 0.900 to 0.930. It contains a Isevo-pinene, 
CioHie, boiling at 160 C. ; thymol, CioH ]4 O, and cymene, Ci Hi 5 , 
boiling at 175 C. 

Oil of wormwood is distilled from the herb Artemisia Absinthium, L. 

Oil of wintergreen is distilled from the leaves of Gaultheria pro- 
cumbens, L. It contains methyl salicylate, C 6 H 4 (OH) COO CH 3 , 
with a little terpene. It is a liquid of pleasant smell and taste, 
boiling at 218 C., and of 1.175 to 1.185 sp. gr. at 15 C. It rotates 
the plane of polarization to the left. It is used as a flavoring essence. 

An artificial oil is made by heating salicylic acid with oil of vitriol 
and methyl alcohol. 


Treatise on the Manufacture of Perfumes. J. H. Snively, New York, 1890. 
Die fliichtigen Oele des Pflanzenreiches. G. Bornemann, Weimar, 1891. 
Handbuch der Parfumerie- und Toilettenseifen-fabrikation. C. Deite, 

Berlin, 1891. (J. Springer.) 

The Art of Perfumery. C. H. Piesse, London, 1891. 5th ed. 
Treatise on the Manufacture of Perfumery. W. T. Brannt, Phila., 1902. 
Odorographia ; a Natural History of Raw Materials and Drugs used in the 

Perfume Industry. J. C. Sawer, London, 1892, Part I. 1894, Part II. 
Perfumes and their Preparation. Askinson-Furst, London, 1892. 
Fabrication des Essences et des Parfums. P. Durvelle, Paris, 1893. 
Descriptive Catalogue of Essential Oils and Organic Chemical Prepara- 
tions. F. B. Power, New York, 1894. (Fritsche Bros.) 
Die Riechstoffe u. Ihre Verwendung. St. Mierzinski, Weimar, 1894. 
Aether und Grundessenzen. Theodor Horatius, Leipzig, 1895. 
Semi-Annual Reports. 1892+. Schimmel and Co. (Fritsche Bi 

Leipzig and New York. 
Huiles essentielles et leurs principaux constituants. Charabot, Dupont 

et Fillet, Paris, 1899. 
Chemistry of Essential Oils and Artificial Perfumes. E. J. Parry, New 

York, 1900. 

Die Aetherischen Oele. F. W. Semmler, Leipzig, 1906. 
Die Aetherischen Oele. Gildemeister und Hoffmann, Berlin, 1910. 
The Volatile Oils. E. Gildemeister and F. Hoffmann. Translated by 

Edward Kremers. New York, 1914. 

jj-f CtJ. Ct~ 


Resins are oxygenated bodies, generally produced by the oxida- 
tion of terpenes or related hydrocarbons in plants or in essential 
oils. They are found as natural or induced exudations from plants, 
often mixed with the essential oil, forming oleo-resin or balsam, or 
with mucilaginous matter, forming gum-resin. True resins are com- 
pact masses, insoluble in water, devoid of marked taste or odor, and 
usually composed of substances of an anhydric or acid nature. They 
are nearly all soluble in alcohol, ether, benzene, and in most vola- 
tile oils, and may usually be saponified with caustic alkali. When 
heated, they soften below their melting points, but cannot be dis- 
tilled undecomposed. The chief uses of resins are : in making var- 
nish ; for soap ; as a constituent of sealing wax ; in medicine, and in 
sizing paper and cloth. 

Common rosin, or colophony, is a resin obtained by the distilla- 
tion of turpentine oil from crude turpentine (p. 388). Three grades 
of rosin are in the market, " virgin," yellow dip, and hard. Virgin 
rosin is made from the first turpentine that exudes after the tree is 
" boxed." It is of a very light yellow or amber color. The greater 
part of the crude turpentine furnishes yellow dip. The hard is made 
from the scrapings from the tree after the turpentine has become 
too thick to run into the box ; it is very dark, being nearly black. 

Rosin is brittle, melts at 100 to 140 G., and has a specific 
gravity of about 1.08. It contains a large amount of abietic anhy- 
dride, C44He2O4, which is readily converted into abietic acid, C^H^Os. 
Rosin is converted by alkalies into " rosin soap " (p. 373), which is 
deliquescent and very soluble in water. Rosin is used as a constitu- 
ent of laundry soaps ; as an addition to cheap varnishes ; as a flux 
in soldering and brazing metals ; in pharmacy ; in ship calking ; and 
as an adulterant of fats, waxes, and mineral oils. 

Rosin must not be confounded with wood-tar, or pitch, obtained 
by the destructive distillation of wood. 

Rosin may be distilled in vacua, or by the aid of superheated 
steam, with very little decomposition ; but when heated in a retort, 
it yields decomposition products consisting of gases, liquids, and 
pitch. The liquid distillate is composed mainly of " rosin spirit," * a 
very complex body, boiling below 360 C. (resembling oil of turpen- 

* Renard. J. Chem. Soc., 46, 843. 


tine, for which it is sometimes substituted), and " rosin oil," * a heav- 
ier liquid, boiling above 360 C. 

The rosin oil is purified by treatment with a little sulphuric acid, 
followed by lime water, and then redistilled, sometimes with caustic 
soda in the still. It has a specific gravity of 0.980 to 1.110 ; is water 
white to brown in color, and is only slightly soluble in alcohol, but 
easily dissolved in fatty oils, ether, chloroform, etc. It is nearly 
odorless, and has a strong, peculiar taste. It is not subject to true 
saponification, although when treated with milk of lime, a combina- 
tion between the terpenes of the oil and the calcium hydroxide takes 
place, forming a solid mass. This is stirred up with more rosin oil, 
to form a soft mixture of about the proportions, 13 CioHi 6 Ca(OH)2, 
which is the commercial " rosin grease/' used as a lubricant on iron 
bearings. Rosin oil is largely used in making such lubricants, and 
as an adulterant for olive and boiled linseed oils. 

Burgundy pitch is a resin resembling common rosin, but obtained 
from the Norway spruce, Picea excelsa, Link. The trees are scari- 
fied, and the resin allowed to harden, when it is collected and treated 
with boiling water, to remove the volatile oils. Its chief constituent 
is abietic anhydride. When stirred up with fats and water, melted 
rosin forms a mass resembling Burgundy pitch in its opacity and 
other properties. 

Mastic and Sandarac are somewhat similar resins, obtained from 
evergreen shrubs which grow along the shores of the Mediterranean 
Sea, especially on the island of Chios, and in northern Africa. The 
former, derived from Pistacia Lentiscus, L., occurs in commerce as 
small translucent grains, or " tears," which soften when masticated, 
and have a slightly bitter, aromatic taste. It is soluble in acetone, 
alcohol, and turpentine oil, and is used in varnish making and in 

Sandarac, also called " gum jumper," is obtained from Callitris 
quadrivalvis, Vent., an evergreen growing in northern Africa. It is 
used in varnishes. 

Amber is a fossil resin found along the coast of the Baltic Sea, in 
Germany. It is the hardest and heaviest of all resins, is capable of 
taking a high polish, and is insoluble in most of the ordinary sol- 
vents. Its color varies from very light yellow to deep brownish red. 
It often contains perfect specimens of fossil insects. When heated 
above its melting point, it is partly decomposed, and then becomes 
soluble in alcohol and in oil of turpentine. 

* Renard. J. Chem. Soc., 24, 304, 1175. 


Transparent pieces of amber are much prized for jewelry, fancy 
articles, mouth-pieces for pipes and cigar holders, and for other orna- 
mental purposes. It is also used in preparing a fine transparent 
varnish for use on negatives in photography. 

When subjected to destructive distillation, amber yields a gas, an 
organic acid (succinic acid), and an oil called " oil of amber." This 
oil and the acid are used somewhat in pharmacy. By treating oil of 
amber with fuming nitric acid, a substance resembling musk in odor 
and other properties is obtained.* 

Copal is a valuable resin. Soft copal, soluble in ether, is obtained 
from living trees in Java, Sumatra, the Philippine Islands, Australia, 
and New Zealand. The better quality, hard copal, is a fossil gum, 
found in irregular lumps, buried in the earth, in the East Indies, 
Madagascar, West Africa, and South America, the last variety being 
called gum animi. Hard copal varies in color from pale yellow to 
brown. Its specific gravity is usually 1.059 to 1.072. It has a higher 
melting point than soft copal, and is insoluble in ether or volatile 
oils. But by heating above its melting point, a partial decomposition 
takes place, and the resin is rendered more soluble in these solvents. 

Hard copal is the hardest of all resins, except amber, and is most 
valuable for varnish making. For this it must first be melted, or 
"run," and while in the liquid state, hot oil of turpentine is slowly 
added and mixed with it. 

Dammar is obtained from a coniferous tree, Agathis loranthifolia, 
Salisb., in the Moluccas. The resin exudes from the tree in drops, 
and is collected after it dries. It is soluble in essential and in fixed 
oils, in crude benzene, and partially so in alcohol and ether. It is 
very light colored, and makes a transparent varnish. 

Kauri, or Australian dammar, is obtained from a New Zealand 
tree, Agathis australis, Stend. Much of the kauri of trade is a fossil 
resin, and is somewhat darker colored than the true dammar and 
copal. It is extensively used for varnish making, being- cheaper 
than copal. 

Dragon's blood is a deep crimson red resin, which exudes from 
the fruit of a palm tree, Dcemonorops Draco, Blume., indigenous in 
the East Indies. It is collected by the natives and made into irregu- 
lar lumps, or cast into long sticks in moulds made by rolling palm 
leaves into cylinders and closing one end. It is freely soluble in 

* The artificial musk of commerce is now made from butyl toluene, by the action 
of nitric and sulphuric acids. Bauer. Berichte der deutschen chemischen Gesell- 
schaft, 24, 2832. 


nearly all of the ordinary solvents, except petroleum ether, oil of 
turpentine, and ether. It is slightly soluble in the two latter. It is 
used in pharmacy, and in certain colored varnishes. 

Guaiacum is a resin derived from certain West Indian trees, 
especially Guaiacum sanctum, L., and G. officinale, L. It exudes from 
the trees through incisions, and forms " tears " or lumps which are 
sent to market. It is soluble in ether, alcohol, chloroform, acetone, 
and caustic soda. Its alcoholic solution is employed as a reagent 
for oxidizing substances, with which it shows a blue color, which is 
destroyed by reducing agents, but reappears when again oxidized. 
Hydrogen peroxide, however, does not change the color to blue unless 
in the presence of blood. Hence guaiacum in alcohol, with hydrogen 
peroxide, is used as a reagent for detecting blood stains. Guaiacum 
is also used in medicine in treating rheumatism and gout. 

Lac is a resin produced by the bite or sting of certain insects, 
Coccus lacca, Kerr, on the small twigs of several species of East In- 
dian trees, of which Ficus Indica, L., and F. religiosa, L., are the 
chief. The resin appears to be formed from the plant sap by the 
female insect, from whose body it exudes, ultimately burying .the 
insect and her eggs, and forming a thick excrescence on the twigs. 
It is collected, together with the twigs which it envelops, and is 
brought into commerce as " stick lac." The insect also secretes a 
brilliant red dye which is extracted by macerating the crude lac in 
warm water. The aqueous solution is evaporated to dryness, and the 
residue sold as lac-dye. After the dye is extracted, the resin is 
known as " seed lac." This is refined by carefully melting and strain- 
ing through muslin bags to remove foreign matter. The melted 
lac is then poured in thin films over cold porcelain, copper, or wood 
cylinders, or plates, and allowed to cool, when it hardens and scales 
off in thin flakes, and is called " shellac." Or it is poured into moulds 
to form " button," or " garnet lac." The shellac is the better quality, 
and is of a pale orange, or red color, and is nearly transparent. It is 
used for spirit varnish. 

Lac is partially soluble in strong alcohol, forming a turbid, gummy 
liquid much used as a varnish and wood filler. It is partly soluble 
in ether, chloroform, and turpentine, but is completely dissolved by 
caustic alkalies and borax solutions. Such solutions are used as 
water varnishes. Lac is also used as the basis of the better grades 
of sealing wax. 

Bleached shellac is made by the action of sodium hypochlorite on 
an alkaline solution of lac ; the lac, precipitated with acid, is melted 


under water and " pulled " to make it white and fibrous. It is used 
for white varnishes, but becomes insoluble in alcohol after some time. 
Elemi is a resin obtained from certain trees, Canarium commune, 
L., in the Philippine Islands, Canarium Mauritianum, Blume., in 
Mauritius, Amyris elemifera in Mexico, and from several varieties of 
Idea in Brazil. The resin varies from white to gray in color, and is 
soft and tough. It softens at 75 C., and melts completely at 120 C. 
It is soluble in alcohol and other solvents, and is used chiefly to give 
toughness to varnishes made from harder resins. 


The resins are chiefly important as furnishing the material for 
varnish making. A varnish is a solution of a resin, or of a drying 
oil, which, when "exposed to the air, becomes hard and impervious to 
air and moisture, through evaporation of the solvent or oxidation of 
the oil. Three classes of varnishes are important : (1) Spirit var- 
nishes, consisting of resin dissolved in alcohol, petroleum spirit, 
acetone, or in any other volatile solvent; (2) turpentine varnishes, 
in which the resin is dissolved in oil of turpentine ; and (3) linseed 
oil varnishes, which may consist of linseed oil alone, or with the 
addition of resin and turpentine oil. 

Spirit varnish dries rapidly, leaving the resin as a thin and bril- 
liant film on the surface to which it is applied. This film is brittle, 
and liable to crack and scale off. The addition of turpentine overcomes 
this difficulty to some extent. Spirit varnishes are often colored with 
dyes soluble in alcohol, or with dragon's blood, gamboge, or cochineal. 
The most important spirit varnishes are made with shellac, though 
mastic, sandarac, and dammar are used. 

Turpentine varnish is tough and flexible, but much slower in drying 
than the spirit varnishes. The resin is simply dissolved in the hot 
oil, and after cooling is ready for use. 

Linseed oil varnishes are the most important. If well-boiled oil 
(p. 357) is applied to a surface, it dries to a hard film, but without 
much brilliancy of surface. By dissolving a resin in the boiled oil 
and thinning to the proper consistency with turpentine, a varnish 
is obtained which dries with a hard, glossy surface, impervious to 
air and moisture. The resins used are mainly amber, copal, anime, 
kauri, and dammar, for transparent varnish. The hard resins are 
not directly soluble in the oil, but must first be partly decomposed 
or " run," by heating above their melting points. There is consider- 
able evolution of irritating gases during this fusion, and an oily 


distillate is often collected. The residue in the pot is then soluble" 
in the hot boiled oil, which is run direct from the boiling kettle into 
the resin melting kettle. After thorough stirring the mixture is 
usually heated some time longer to secure homogeneous solution. It 
is then cooled to about 130 or 140 C., and thinned to the desired 
consistency with oil of turpentine. The varnish is allowed to stand 
in storage tanks for several months, or even for a year or two, until 
thoroughly clarified. 

The boiling of the oil and of the varnish involves considerable risk 
from fire. The oil froths very much, and the vapors given off are 
inflammable, hence it is usually the custom to build the furnace with 
the fire-door opening through a partition into another room. The 
vapors should be led into a flue having a good draught. 


Oleo-resins are mixtures of the resin and the essential oil of the 
plant from which they exude. Among them is a group of substances 
which have peculiar odor and pungent taste, and which are called 
balsams. They are the exudations from tropical trees belonging to 
the genera Myroxylon and Styrax. The most important are Benzoin, 
Peru, Tolu, and Storax balsams. They contain free benzoic or cin- 
namic acids, or compounds of them, to which their peculiar properties 
are due. The balsams are chiefly used in medicine and pharmacy, 
and for incense and perfumes. 

The so-called Canada balsam is an oleo-resin containing turpen- 
tine, and is not a true balsam. 


Gum resins are exudations from plants ; they are the inspissated 
juice, and contain both gum and resin. They form emulsions with 
water, a portion of the gum dissolving. 

Ammoniacum is derived from a Persian plant, Dorema Ammonia- 
cum, Don. It forms drops, yellow on the surface and milky within. 
It is partly soluble in water, and has a peculiar odor and bitter taste. 
It is employed in medicine. 

Asafoetida is obtained from the roots of two plants, Ferula Nar- 
thex, Boiss., and F. fatida, Regel, native in Thibet and Turkistan. 
It forms tears' and nodules, frequently contaminated with earthy 
impurities. It has a powerful garlic odor and bitter taste. It is 
mainly used in medicine as a stimulant. 


Euphorbium is derived from a species of cactus, Euphorbia resini- 
fera, Berg., native in Morocco. It has a very pungent taste, an aro- 
matic odor, and the powdered gum irritates the throat and nose. 
It is a violent emetic and purgative, and is chiefly used in veteri- 
nary medicine. 

Galbanum is obtained from Persian plants, probably Ferula gal- 
baniflua, Boiss. & Buhse. It forms tears, or irregular lumps, of 
brownish yellow color, aromatic odor, and bitter taste. The several 
varieties found in commerce are used in medicine, and as constit- 
uents of incense. 

Gamboge (p. 241) is an orange-red substance, derived from a tree, 
Garcinia Hanburyi, Hook., or G. Morella, Desr., native in Cochin 
China and Siam. It is soluble in alcohol, has an acrid taste, and is 
a powerful purgative. Its chief uses are in medicine, and as a 

Myrrh is obtained from a shrub, Commiphora Myrrha, Engl., 
growing on the coast of Arabia. It comes in commerce as red- 
brown, dusty lumps, breaking with an oily-appearing fracture. It 
has a fragrant odor and bitter taste, and emulsifies with water. 
It is used as a tonic in medicine, and in preparing incense. 

Olibanum or frankincense is derived from several species of Bos- 
wellia, the trees being native in Africa and Arabia. It forms tears 
of a yellow-brown color and milky appearance. It has a slight 
turpentine-like taste, and an aromatic odor. It forms an emulsion 
with water, and was formerly much used in medicine. It is now 
chiefly employed in preparing incense. 


Gums are amorphous bodies of complex constitution, nearly all of 
vegetable origin, and soluble in, or, at least, gelatinizing with water, 
but insoluble in alcohol. When boiled with dilute acid, they yield 
sugars, and when oxidized are converted into oxalic or mucic acids. 

Acacia, Gum Arabic, or Gum Senegal, is derived from numer- 
ous plants of the Acacia family, mostly native in Africa. It forms 
lumps of various sizes, ranging in color from transparent white to 
red-brown. Its chief constituent is arabic acid, or arabin, C^H^On, 
as calcium salt. It dissolves in cold or hot water with equal readi- 
ness, and is much used in pharmacy in preparing emulsions. Low 
grades are used for mucilage, in calico printing, in thickening ink 
and water colors, and as stiffening in cloth. 


Tragacanth is an exudation from Astragalus gummifer, Labill., 
growing in the Levant. It forms dull white, translucent plates, 
which swell in water and partly dissolve, forming a thick mucilage. 
Its uses are similar to those of gum arabic. 

Agar-agar or Bengal isinglass is a dried seaweed, Gracilaria liche- 
noides and Eucheuma spinosum, collected in China. It forms a jelly 
with water. 

Iceland moss, Cetraria islandica, yields a jelly containing two 
gums, lichenine, CeHioOs, and wolichenine. The former is not colored 
blue by iodine, while the latter is. 

Irish moss, Chondrus crispm, yields a soluble gum, which is not 
colored blue by iodine. 


Report on the Gums, Resins, Oleo-resins, and Resinous Products of India. 

M. C. Cooke, London, 1874. 

Varnishes, Lacquers, Siccatives, and Sealing Waxes. E. Andres. Trans- 
lated by Wm. T. Brannt, Philadelphia, 1882. (H. C. Baird & Son.) 
Oils and Varnishes. James Cameron, Philadelphia, 1886. (Blakiston, 

Son & Co.) 

Der Fabrikation der Lacke und Firnisse. Paul Lohmann, Berlin, 1890. 
Fossil Resins. C. Lawn and H. Booth, New York, 1891. 
Die Fabrikation der Lacke Firnisse, u. s. w. E. Andres. 4 te Auf. Wien, 


Notes on Varnish and Fossil Resins. R. I. Clark, London, 1892 (?). 
Painters' Colours, Oils, and Varnishes. G. H. Hurst, London, 1892. 
The Chemistry of Paints and Painting. A. H. Church. 2d ed. London, 


Pigments, Paints, and Painting. G. Terry, London, 1893. (Spon & Co.) 
Fabrication des Vernis. L. Naudin, Paris, 1893. 
Die Fabrikation der Copal-, Turpentinol- und Spiritus-Lacke. L. E. 

Andes. 2 te Auf. Wien, 1895. (Hartleben.) 
Couleurs et Vernis. G. Halphen, Paris, 1895. 

Die Harze und ihre Producte. G. Thenius, Wien, 1895. (Hartleben.) 
Gummi arabicum u. dessen Surrogate in festem u. niissigem Zustande. 

L. E, Andes, Wien, 1896. (Hartleben.) 

Die Aetherischen Oele. Gildemeister und Hoffmann, Berlin, 1899. 
The Chemistry of Essential Oils and Artificial Perfumes. E. J. Parry, New 

York, 1900. 


Starch is widely and abundantly distributed in the vegetable 
kingdom, occurring in nearly all plants in a greater or less quantity. 
It forms rounded grains of characteristic appearance in the several 
varieties, and is most abundant in the fruit, tubers, seeds, and stems 
of the plants from which it is industrially obtained. It is a typical 
carbohydrate, and on analysis corresponds to the formula CeHioOs; 
but it is probable that the true symbol is some multiple of this, and 
that the formula should be written (CeHioOs)^ where n is 4 or more. 
Starch has not yet been prepared synthetically, and even its for- 
mation in plants is not fully understood ; but it appears that the 
chlorophyl (the green coloring matter in plants) enters into the 
reaction in some way, perhaps as a " contact " substance. The 
carbon dioxide of the air is reduced by the joint action of the chloro- 
phyl and sunlight, the carbon being assimilated, and part of the 
oxygen, at least, being set free. The formation of starch might be 
represented thus : 

6 CO 2 + 5 H 2 O = C 6 H 10 O 5 + 6 O 2 . 

It is, however, probably not formed directly, but may be an altera- 
tion product of the sugar which is so formed. As hypothetical 
reactions, the following will serve to show the outline of the process, 
but it is by no means certain that these truly represent the exact 
changes which occur : 

6 CO 2 + 6 H 2 O = C 6 Hi 2 O 6 + 6 O 2 . 
CeH^Oe = C 6 Hi O 5 + H 2 O. 

It appears somewhat improbable that substances of such high molec- 
ular weight as glucose, C 6 Hi 2 O 6 , or starch, should be formed directly 
from the reduction of carbon dioxide. According to Baeyer,* it is 
more probable that formaldehyde, CH 2 O, is first produced, and 
then by a polymerization process, the glucose is formed, from which 
starch is derived : 

6 CO 2 -f 6 H 2 O = 6 CH 2 O + 6 O 2 . 
6 CH 2 O = C 6 Hi 2 O 6 . 

The starch is formed in the leaves and green parts of the plant, 
being then transported in soluble form to the other parts, where it is 

* Berichte der deutschen chemischen Gesellschaft, 3, 67. 
2o 401 


at once applied to the building up of the tissues, or is deposited as 
reserve material for the future nourishment of the plant, or of a new 
individual; the greatest deposits are generally found in the roots, 
tubers, or seeds. 

As seen under the microscope, a starch granule is made up of 
different layers, arranged around a nucleus, a dark interior portion, 
generally at one side of the granule. Each granule consists of an 
interior substance called " granulose," and an exterior transparent 
covering, inert and insoluble, and resembling cellulose in structure. 
But recent investigations tend to prove that the " starch cellulose " 
is not present as such in the granule, but is formed from the starch 
substance by the action of acids or by fermentation. 

Starch is entirely insoluble in cold- water, but when heated to 70 
or 80 C., the granules swell and finally burst, and the starch sub- 
stance, "granulose," combines with the water to form paste. When 
this is boiled in an excess of water, it goes into solution and may be 
filtered. The solution yields an intense blue color with iodine, hence 
its use as an " indicator " ; it is optically active and rotates the plane 
of polarization tr the left. 

By exposing starch to the action of cold dilute mineral acid for 
several days, it is converted into a soluble modification called amylo- 
dextrin, which dissolves in warm water without forming a paste. 
When heated dry to 200 C., starch is converted into dextrine or 
British gum. 

The chief industrial sources of starch are potatoes, wheat, corn, 
rice, arrowroot, and certain varieties of palm trees (sago). In Europe, 
potatoes, rice, and wheat are used, while in this country corn and 
wheat are mainly employed. The separation of the starch, which is 
mixed with various nitrogenous and fatty matters and some mineral 
impurities, is essentially a mechanical process ; but much care is 
needed to prevent changes which would spoil the product. 

Corn starch * is usually made by the alkaline or " sweet " process ; 
sometimes by an acid or fermentation method similar to that em- 
ployed for wheat starch. In the alkaline process the grain is run 
through a fanning mill to blow away dust, husks, etc., and is then 
steeped in water at from 70 to 140 F. for from three to ten days, 
when the softened grains are crushed between rolls. This steeping 
removes much of the oil and swells the gluten and albuminous matter 
so that it is readily attacked by the alkali. After a time putrefactive 
fermentation sets in and hydrogen sulphide is evolved. Since this 
* J. Soc. Chem. Ind., 1887, 80. 1902, 4. Geo. Archbold. 


causes a nuisance, tfie method has been replaced in some factories 
by the Durgen system, in which a continuous stream of water at 
130 to 140 F. flows slowly through the steeping tanks. After 
three days the grain is soft, while a large quantity of extractive 
matter has been washed away. The grain is then ground in buhr- 
stone and roller mills through which water is flowing; the starchy 
magma goes to revolving sieves of brass wire for the coarser strain- 
ing, and then to cylindrical reels covered with bolting cloth. The 
mass which passes over the sieves is reground and again sifted. 
The waste glutinous matter is pressed and dried for cattle feeding, 
or is sold wet as " swill " for hogs. 

The milky liquor from the sieves is settled and drawn off from 
the crude starch, which is washed twice with fresh water and then 
pumped into vats having good stirring apparatus, and provided with 
holes in the sides, closed by plugs and used for decanting the liquor. 
A dilute caustic soda solution of 7 or 8 Be. is stirred into the starch 
until the liquid becomes greenish yellow ; then the whole is stirred 
for several hours. When a test shows that the suspended matter 
settles in two layers, the starch on top, sedimentation is allowed to 
take place and the supernatant liquor, containing much oil and 
nitrogenous matter in solution, is drawn off. The sediment is stirred 
up with water, allowed to stand until the gluten has deposited, 
and then, by pulling the plugs in succession, the starch in suspen- 
sion is " siphoned off " into tanks. By several repetitions of this 
process the starch is nearly all removed from the gluten and at the 
same time is separated into several grades. The residue then flows 
on to a long, slightly inclined table, or " run," from 60 to 120 feet 
long and having a fall of 3 or 4 inches. A stream of water flows 
slowly over it and washes away the gluten and fibrous matter, while 
the starch deposits on the table. 

The starch collected in the several tanks is washed with water 
and sometimes again siphoned, and is then run through bolting 
cloth to the settling tanks, where it deposits in a dense compact 
layer from which the water can be drawn off. The wet starch is then 
shovelled into frames lined with cloth .and having perforated bottoms, 
through which the water drains. The cake of damp starch is cut 
into smaller blocks and placed on porous floors of plaster of Paris 
or brick, which absorb the adhering water.* The starch is removed 
to the dry room and kept at a temperature of 125 F. for several 

* These floors may be subsequently dried by passing hot air through flues ar- 
ranged in them. 


days. While it is drying, the impurities still remaining in it find 
their way to the surface, where they form a yellowish deposit which 
is cut away when the starch is nearly dry. The block is then wrapped 
in paper and further dried at 150 to 170 F. for several days. Dur- 
ing this time the mass contracts and cracks into a number of irreg- 
ularly shaped prismatic rods, called " crystals," though they are not 
true crystals. The entire drying process requires several weeks, and 
the product as sent to market contains about 10 to 12 per cent of water. 

An improved process is now used as follows : The shelled corn is 
screened to remove dirt, husks, etc. ; it then passes magnets to re- 
move nails or bits of iron, and then goes into wooden " steep-tanks " 
of 1000 to 1500 bushels capacity. Each tank has a false bottom, and 
a circulation pipe on the outside, passing from the false bottom to near 
the top of the tank. Steam can be injected into the pipe to maintain 
the circulation and keep the steep water at about 60 C. Steeping 
continues 24 hours or more, and the steep liquor contains about 0.3 
per cent sulphurous acid to prevent fermentation. The liquor is drawn 
off and the softened grain crushed in " cracker " mills to loosen the 
germs. These mills are large disks, set face to face, having projecting 
teeth and rotating in opposite directions. 

The coarse meal passes to " separators," long, narrow tanks, con- 
taining a starch milk of 10 to 12 Be., calcium chloride being some- 
times dissolved in the liquor to increase its density. The germs, being 
light, float over the dam at the end of the tank, while the hulls and 
starchy portions sink, and pass out by an opening at the bottom of 
the tank. The germs pass over copper screens or " shakers," where 
they are sprinkled with water to free them from adhering starch ; the 
starch milk thus obtained is returned to the separator. 

The germs are pressed to remove water, dried, and ground fine; 
the meal is heated and heavily pressed in a hydraulic press (p. 353) 
to obtain the corn oil ; the oil cake is sold for cattle food. 

The hulls and starchy matter from the separators are ground fine 
in buhrstone mills and passed over copper "shakers," some of the starch 
milk going back to the separators, and the rest passing to shakers 
covered with silk bolting cloth ; the chaff and husks are reground and 
passed through a slop machine to remove the last portion of starch. 

The starch liquors, containing gluten and other substances, are 
agitated in a mixing tank with dilute caustic soda solution ; this dis- 
solves some of the gluten, swells the remainder, neutralizes the acid, 
and coagulates the fine suspended impurities. The magma then 
goes to the " runs," or " table," where the starch deposits, the lighter 


gluten being washec| into a settling tank, from which it is pumped 
into a filter-press to remove the water. The gluten is then dried 
and sold as sucli, or is pulverized and mixed with the bran and husks 
from the slop machine. The steep water from the softening of the 
grain carries considerable soluble matter and is evaporated to about 
30 Be., and mixed with the bran and husks before drying. 

The " green starch " from the tables is usually mixed with water 
and again passed over the tables, when dry starch is to be made. 

Centrifugal machines are sometimes used for separating the 
starch from the wash water. These machines are of two kinds, 
those having a perforated basket, and those in which the basket 
is of unperf orated sheet metal. In the latter, the starch is thrown 
against the cylinder wall and packed so firmly that it remains as 
a thick layer, while the water collects in the middle of the drum 
and can be drawn off very completely, carrying with it much of the 
glutinous and fibrous matter. In a perforated drum the water 
passes through, leaving the solid matter behind. The starch, being 
heavier than the cellulose, forms a layer directly on the basket walls, 
while inside of this is a layer of gray starch containing the impuri- 
ties ; this latter is scraped off and washed again. The starchy 
liquid running into the basket must not be too thick, otherwise the 
load does not distribute itself evenly in the basket. 

Corn contains about 54 per cent of starch, and the actual yield 
obtained in technical work is about 50 per cent, or 28 pounds of 
starch from a bushel (56 pounds) of corn. About 13 pounds of glu- 
ten suitable for cattle food is also recovered per bushel of corn. 

The best grade of corn starch is largely consumed for food, but its 
principal use is in laundry work. Lower grades are chiefly employed 
in manufacturing and in textile industries. In some technical opera- 
tions the so-called " green starch " is used. This is the product 
obtained directly from the inclined table, settling tanks, or centrifu- 
gal machine after a partial drying. It contains some impurities 
and is generally damp, often containing 40 per cent of water. It is 
mainly employed for glucose making, for stiffening and size, in color 
mixing for calico printing, and in the manufacture of paper boxes. 

The old fermentation processes of starch extraction destroy the 
gluten and cause incipient hydrolysis of the product ; the paste made 
from such starch is more limpid than that of starch made by newer 
methods in which these changes do not occur. In modern work, 
to obtain this quality of " thin boiling " paste, the starch, after 
separation from the gluten, is given a mild treatment with mineral 



acid at temperatures of 26 to 40 C. Thin boiling starch is preferred 
for textile work because oi the greater fluidity and better penetra- 
tion of the paste into the fabric. 



(A) Steeped 

() Crushed 

(C) Germ separators 







Corn meal 


Ground fine (buhrstones) 


Passed twice over shakers which 
are sprinkled with water 

Oil cake 

Corn oil 

(E) Starch water 

Carries starch and gluten 

Agitated in tank 

Fed to "runs" or tables 
(120 ft. by 2 ft., with 
gentle incline) 

(D) Husks 

Reground (steel rolls) 


"Slop" passed through 
slop machine, a wringer 
to remove residual starch 

Starch water Wet feed 



Starch to (E) 





Cake ground and passed 
through driers 

Consists of 

husk and bits 

of germ 

Passed to dry 



Dry feed 

Process generally repeated 




Wheat starch is made by the fermentation or "sour " process, or 
by Martin's process without fermentation. 

By the sour process, all the gluten, of which wheat contains a 
large amount, is destroyed, consequently there is considerable loss. 
The grain is soaked in water until soft, and then crushed between 
rolls or pressed in bags. The starch is washed out of the crushed 
pulp with water, and the milky liquid is run into tanks and allowed 
to ferment. In order to hasten this, some of the sour liquor from a 
previous fermentation is added. The temperature is kept at about 
20 C., and the contents of the cistern well stirred frequently. The 
fermentation lasts from 10 to 14 days; the sugar, albumin, and 
gummy matters of the wheat undergo an alcoholic fermentation, 
followed by the development of acetic, lactic, and butyric acids. 
These acids then attack the gluten, dissolving it in part, and de- 
stroy its tough and sticky properties, so that it is easily washed 
free from the starch. The washing is done in revolving sieves, in 
which the swollen gluten, cellulose, etc., remain. The starch is re- 
peatedly washed and sieved, or levigated, until sufficiently pure and 
white, when it is dried as already described under corn starch; but 
more care is necessary, because of the tendency of the mass to cake 
together, owing to the presence of a trace of gluten. The process 
must be carefully watched lest the fermentation go too far and putre- 
faction set in, thus causing a loss of starch. The acid waste liquors 
are difficult to dispose of, and cause considerable nuisance in the 
neighborhood. Usually about 59 pounds of starch and 11 pounds 
of bran are obtained from 100 pounds of wheat. But only a small 
quantity of sour gluten is recovered. 

By Martin's process part of the nitrogenous matter (gluten) is 
recovered. Ordinary wheat flour from which the bran has been 
removed is used instead of the whole grain. The flour is kneaded 
with 40 per cent of water to form a stiff dough, which is then washed 
in small portions at a time in a fine sieve, while small jets of water 
continually play upon the mass, carrying away the starch. By 
treating the partly washed starch with a solution of caustic soda 
(sp. gr. 1.013) and allowing it to stand a few hours, the remaining 
gluten is swollen and may be removed by sieving on bolting cloth. 
The pasty mass of gluten left in the sieves is utilized in the manu- 
facture of macaroni, noodles, and gluten bread, but more especially 
for paste and for cement for leather, and as a thickening material 
instead of casein or albumin in textile working. 

Fesca's modification of Martin's process consists in stirring wheat 


flour into water to form a thin " milk," which is then run into cen- 
trifugal machines. The starch, being heavier than the gluten, col- 
lects next the revolving sieve. The interior layer, consisting of a 
mixture of starch and gluten, is removed, washed with water, and 
again " centriffed." But much starch remains in the gluten. 

The yield by Martin's method is about 55 pounds of starch and 
12 pounds of gluten from 100 pounds of wheat. By Fesca's process, 
only 40 to 45 pounds of starch are obtained from 100 pounds of wheat. 

Potato starch is most important in Europe. The tubers contain 
an average of about 20 per cent of starch and 75 per cent water. 
The skin contains some fats and coloring matter, but no starch. 
The adhering dirt and sand are carefully removed by washing in 
a revolving drum made of wood or iron slats with narrow openings 
between them for the escape of the dirt, etc. Inside the drum are 
revolving arms which rub the potatoes together, or revolving wire 
or bristle brushes which scrub them as the drum turns. The wash- 
ing must be thorough or the quality of the starch suffers. The tubers 
are next rasped or ground in a machine consisting of a revolving 
cylinder or roll, around whose outer surface are set a large number 
of narrow knife-edges or saw-blades, which project about one-fifth 
of an inch. These knife-edges rotate very close to fixed wooden bars 
which catch and hold the potato while it is scraped into soft pulp. 

The starch in the potato is enclosed in little cells or bags of cellu- 
lose, a number of granules being in each cell. Since the starch can 
only be washed away from the ruptured cells, the finer the pulp the 
larger the yield of starch. But even with the best raspers many 
cells escape unbroken, and usually about 15 per cent of the starch is 
lost. Sometimes the pulp is reground after it has been washed, 
which increases the yield of starch slightly. 

The pulp, consisting of starch and cellulose fibre and tissue, 
passes into a series of shaking sieves, where the starch is washed 
away with a limited amount of water. A better apparatus consists 
of a series of revolving wire gauze cylinders (30 to 35 meshes to the 
inch), containing brushes which revolve in a direction opposite to 
the motion of the cylinder. Fine jets of water play upon the pulp 
and wash out the starch. The milky liquor passes to a revolving 
sieve with 50 meshes per inch, which retains any fibre that passes 
through the coarser screens. Long semicylindrical sieves contain- 
ing brushes set in the form of an Archimedian screw around a revolving 
shaft are sometimes used. The brushes push the pulp along from one 


end to the other, at -the same time thoroughly working it over, while 
the starch is washed out by jets of water. The waste pulp passing 
over the sieves is treated by Biittner and Meyer's process; it is 
pressed and dried rapidly until the moisture is about 12 per cent. 
It is sold as a low-grade cattle food. 

The starch suspended in the wash water is run over inclined tables 
similar to those already described. The crude product is stirred up 
with water in a tank, and after the sand and heavy dirt have settled, 
the starch in suspension is rapidly " siphoned " off through holes in 
the side of the tank. By levigation, the starch is obtained in several 
grades of purity. Centrifugal machines are also employed to separate 
the starch and wash water, but with less success than in the case of 
corn, wheat, or rice starch. 

The crude starch obtained by any of 'these methods is purified 
by repeated washings and levigation, with an occasional passing 
through sieves or bolting cloth to remove fibre. The purified starch 
is dried in much the same way as is corn starch. 

Potato starch is also made by the " rotting " process, in which 
the moist, sliced material is heaped in a warm room. Fermentation 
and ultimate decomposition of the cell walls takes place, so that the 
starch can be washed out of the pulp. Much care is necessary that 
the fermentation does not attack the starch itself. The mass must 
be turned over frequently during decomposition. 

The wash waters from potato starch contain much potash, phos- 
phoric acid, albumin, and nitrogenous matter, which soon ferment 
and become very offensive. If possible, they should be used at once 
to irrigate land. Much ingenuity has been expended to devise means 
of making them less offensive, but without much success. 

The yield from 100 pounds of potatoes is about 15 or 16 pounds 
of dry starch. The product is chiefly used in the textile industries, 
for laundry purposes, and in glucose and dextrine making ; for the 
two last mentioned it is customary to use the " green starch," con- 
taining from 30 to 40 per cent water. 

Rice starch * is chiefly produced in Europe, only the broken 
grains separated with the husks in the cleansing mills being used. 
Rice contains nearly 80 per cent of starch, but its separation is diffi- 
cult, since the cells of the grain are composed of dense glutinous 
material and the starch granules are cemented together solidly by 
albuminous and gummy matter. In order to soften the gluten, the 
rice is macerated in dilute caustic soda (sp. gr. 1.007) containing 
* J. Berger, Chem. Zeitung, 14, 1440 and 1571 ; 15, 843. 


about 0.5 per cent caustic. After soaking about 18 hours with fre- 
quent stirring, the liquor is drawn off and a fresh caustic solu- 
tion is added. When the grain is soft, it is crushed in mills while 
a stream of dilute caustic plays over the mass, dissolving a part of 
the glutinous matter, and swelling the remainder so that it, together 
with the fibrous matter, may be removed by sieving. The starch 
is then separated from the liquor by centrifugal machines, and further 
purified by washing it with water, running it through centrifugal ma- 
chines or settling tanks, and finally filter-pressing and dr